ST2195 Coursework
Celest Lee
2022-04-01
This RMD file contains all coursework questions. Disclaimer: Some graphs might not be well-scaled (especially grid plots)
Setting working directory and loading relevant libraries
setwd("~/Desktop/data_files/dataverse_files")# install.packages("ggplot2")
# install.packages("tidyr")
# install.packages("ragg")
# install.packages("cowplot")
# install.packages("dplyr")
# install.packages("reshape2")
# install.packages("scales")
# install.packages("lubridate)
# install.packages("future")
# install.packages("mlr3")
# install.packages("mlr3verse")
# install.packages("ranger")
library(ggplot2)
library(tidyr)
library(ragg)
library(cowplot) # For grid plots
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(reshape2) # To melt and cast dataframes##
## Attaching package: 'reshape2'## The following object is masked from 'package:tidyr':
##
## smithslibrary(scales) # For axis labels
library(lubridate) # To convert time##
## Attaching package: 'lubridate'## The following object is masked from 'package:cowplot':
##
## stamp## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, union# Libraries for machine learning
library(future)
library(mlr3verse)## Loading required package: mlr3library(mlr3learners)
library(mlr3pipelines)
library(mlr3tuning)## Loading required package: paradoxlibrary(mlr3viz)
library(paradox)
library(glmnet)## Loading required package: Matrix##
## Attaching package: 'Matrix'## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpack## Loaded glmnet 4.1-3library(ranger)
future::plan()## sequential:
## - args: function (..., envir = parent.frame())
## - tweaked: FALSE
## - call: NULLLoading data files – Years 2005 to 2007 and binding it into data
frame years Loading supplementary datasets –
planes and airports
# Creating tables saved in csv files
planes <- read.csv("/Users/celestlee/Desktop/data_files/dataverse_files/plane-data.csv", header = TRUE)
airports <- read.csv("/Users/celestlee/Desktop/data_files/dataverse_files/airports.csv", header = TRUE)
# Might take some time to load!
years <- rbind(read.csv("/Users/celestlee/Desktop/data_files/dataverse_files/2005.csv.bz2"), read.csv("/Users/celestlee/Desktop/data_files/dataverse_files/2006.csv.bz2"), read.csv("/Users/celestlee/Desktop/data_files/dataverse_files/2007.csv.bz2"))Summary of data frame years
str(years)## 'data.frame': 21735733 obs. of 29 variables:
## $ Year : int 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005 ...
## $ Month : int 1 1 1 1 1 1 1 1 1 1 ...
## $ DayofMonth : int 28 29 30 31 2 3 4 5 6 7 ...
## $ DayOfWeek : int 5 6 7 1 7 1 2 3 4 5 ...
## $ DepTime : int 1603 1559 1603 1556 1934 2042 2046 NA 2110 1859 ...
## $ CRSDepTime : int 1605 1605 1610 1605 1900 1900 1900 1900 1900 1900 ...
## $ ArrTime : int 1741 1736 1741 1726 2235 9 2357 NA 8 2235 ...
## $ CRSArrTime : int 1759 1759 1805 1759 2232 2232 2232 2232 2223 2223 ...
## $ UniqueCarrier : chr "UA" "UA" "UA" "UA" ...
## $ FlightNum : int 541 541 541 541 542 542 542 542 542 542 ...
## $ TailNum : chr "N935UA" "N941UA" "N342UA" "N326UA" ...
## $ ActualElapsedTime: int 158 157 158 150 121 147 131 NA 118 156 ...
## $ CRSElapsedTime : int 174 174 175 174 152 152 152 152 143 143 ...
## $ AirTime : int 131 136 131 129 106 97 100 NA 101 96 ...
## $ ArrDelay : int -18 -23 -24 -33 3 97 85 NA 105 12 ...
## $ DepDelay : int -2 -6 -7 -9 34 102 106 NA 130 -1 ...
## $ Origin : chr "BOS" "BOS" "BOS" "BOS" ...
## $ Dest : chr "ORD" "ORD" "ORD" "ORD" ...
## $ Distance : int 867 867 867 867 867 867 867 867 867 867 ...
## $ TaxiIn : int 4 6 9 11 5 3 5 0 2 4 ...
## $ TaxiOut : int 23 15 18 10 10 47 26 0 15 56 ...
## $ Cancelled : int 0 0 0 0 0 0 0 1 0 0 ...
## $ CancellationCode : chr "" "" "" "" ...
## $ Diverted : int 0 0 0 0 0 0 0 0 0 0 ...
## $ CarrierDelay : int 0 0 0 0 0 23 46 0 16 0 ...
## $ WeatherDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ NASDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ SecurityDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ LateAircraftDelay: int 0 0 0 0 0 74 39 0 89 0 ...summary(years)## Year Month DayofMonth DayOfWeek
## Min. :2005 Min. : 1.00 Min. : 1.00 Min. :1.000
## 1st Qu.:2005 1st Qu.: 4.00 1st Qu.: 8.00 1st Qu.:2.000
## Median :2006 Median : 7.00 Median :16.00 Median :4.000
## Mean :2006 Mean : 6.52 Mean :15.73 Mean :3.943
## 3rd Qu.:2007 3rd Qu.: 9.00 3rd Qu.:23.00 3rd Qu.:6.000
## Max. :2007 Max. :12.00 Max. :31.00 Max. :7.000
##
## DepTime CRSDepTime ArrTime CRSArrTime
## Min. : 1 Min. : 0 Min. : 1 Min. : 0
## 1st Qu.: 931 1st Qu.: 930 1st Qu.:1110 1st Qu.:1116
## Median :1330 Median :1325 Median :1516 Median :1520
## Mean :1341 Mean :1334 Mean :1487 Mean :1497
## 3rd Qu.:1733 3rd Qu.:1721 3rd Qu.:1912 3rd Qu.:1909
## Max. :2930 Max. :2359 Max. :2955 Max. :2400
## NA's :416412 NA's :463805
## UniqueCarrier FlightNum TailNum ActualElapsedTime
## Length:21735733 Min. : 1 Length:21735733 Min. : -66.0
## Class :character 1st Qu.: 586 Class :character 1st Qu.: 75.0
## Mode :character Median :1481 Mode :character Median : 107.0
## Mean :2140 Mean : 125.7
## 3rd Qu.:3364 3rd Qu.: 156.0
## Max. :9619 Max. :1879.0
## NA's :463805
## CRSElapsedTime AirTime ArrDelay DepDelay
## Min. :-1240.0 Min. :-1428.0 Min. :-939.0 Min. :-1200.0
## 1st Qu.: 76.0 1st Qu.: 54.0 1st Qu.: -9.0 1st Qu.: -4.0
## Median : 108.0 Median : 84.0 Median : -1.0 Median : 0.0
## Mean : 126.8 Mean : 102.3 Mean : 8.7 Mean : 10.1
## 3rd Qu.: 156.0 3rd Qu.: 131.0 3rd Qu.: 13.0 3rd Qu.: 9.0
## Max. : 1430.0 Max. : 1958.0 Max. :2598.0 Max. : 2601.0
## NA's :998 NA's :463805 NA's :463805 NA's :416412
## Origin Dest Distance TaxiIn
## Length:21735733 Length:21735733 Min. : 11.0 Min. : 0.000
## Class :character Class :character 1st Qu.: 317.0 1st Qu.: 4.000
## Mode :character Mode :character Median : 569.0 Median : 5.000
## Mean : 723.8 Mean : 7.101
## 3rd Qu.: 950.0 3rd Qu.: 8.000
## Max. :4962.0 Max. :1523.000
##
## TaxiOut Cancelled CancellationCode Diverted
## Min. : 0.00 Min. :0.00000 Length:21735733 Min. :0.00000
## 1st Qu.: 10.00 1st Qu.:0.00000 Class :character 1st Qu.:0.00000
## Median : 13.00 Median :0.00000 Mode :character Median :0.00000
## Mean : 15.83 Mean :0.01916 Mean :0.00218
## 3rd Qu.: 19.00 3rd Qu.:0.00000 3rd Qu.:0.00000
## Max. :1339.00 Max. :1.00000 Max. :1.00000
##
## CarrierDelay WeatherDelay NASDelay SecurityDelay
## Min. : 0.000 Min. : 0.0000 Min. : -49.000 Min. : 0.0000
## 1st Qu.: 0.000 1st Qu.: 0.0000 1st Qu.: 0.000 1st Qu.: 0.0000
## Median : 0.000 Median : 0.0000 Median : 0.000 Median : 0.0000
## Mean : 3.428 Mean : 0.7042 Mean : 3.582 Mean : 0.0248
## 3rd Qu.: 0.000 3rd Qu.: 0.0000 3rd Qu.: 0.000 3rd Qu.: 0.0000
## Max. :2580.000 Max. :1510.0000 Max. :1392.000 Max. :382.0000
##
## LateAircraftDelay
## Min. : 0.000
## 1st Qu.: 0.000
## Median : 0.000
## Mean : 4.436
## 3rd Qu.: 0.000
## Max. :1366.000
## 1. When is the best time of day, day of the week, and time of year to fly to minimise delays?
Data wrangling for years data frame and creating
variable status
Setting Month, Year, DayOfWeek
and DayofMonth as factors in data frame
years
years$Month <- as.factor(years$Month)
years$Year <- as.factor(years$Year)
years$DayOfWeek <- as.factor(years$DayOfWeek)
years$DayofMonth <- as.factor(years$DayofMonth) Adding new column status to years for
flight status
years$status <- NA
years$status[years$Diverted == 1] <- "Diverted"
years$status[years$DepDelay > 0] <- "Delayed"
years$status[years$Cancelled == 1] <- "Cancelled"
years$status <- ifelse(years$Diverted != 1 & years$DepDelay <= 0 &
years$Cancelled != 1, "On Time", years$status)
years$status <- as.factor(years$status)Summary of column status in data frame
years
summary(years$status)## Cancelled Delayed Diverted On Time
## 416412 8508330 21996 12788995Bar plot of status of flights in percentage
# Creating data frame `status_perc` with `perc` as percentages of `status`
status_perc <- years %>%
count(status) %>%
mutate(perc = n / nrow(years) * 100)
# Creating plot labels for `perc`
perc_label <- c("0", "20", "40", "60")
# Bar plot
status_perc %>%
ggplot(aes(x = status, y = perc)) +
geom_bar(stat = "identity", fill = "steelblue") +
scale_y_continuous(labels = paste0(perc_label, "%")) +
geom_text(aes(label = paste0(round(perc,2), "%")), position = position_dodge(width = 0.9), vjust = -0.4) +
labs(title = "Percentage of flight status between 2005 to 2007", x = "Flight Status", y = "Percentage") +
theme_bw() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12)) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) 
Delay rate by month (Line plot)
# Creating plot labels for `Month`
month_label <- c("Jan","Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov","Dec")
# Creating data frame `month_delay` with variable `delay_rate` to calculate delay rate by month
month_delay <- years %>%
select(Month, status) %>%
group_by(Month) %>%
summarize(num_delays = sum(status == 'Delayed'),
num_flights = n(),
delay_rate = sum(status == 'Delayed') / n())
# Line plot of **delay rate by month**
plot_month <- month_delay %>%
ggplot(aes(x = Month, y = delay_rate, group = 1)) +
geom_line(color = "steelblue") +
geom_point(color = "steelblue") +
scale_x_discrete(labels = month_label) +
labs(title = "Month", y = "Delay Rate") +
theme_classic() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
plot_month
Delay rate by day of month (Line plot)
# Creating data frame `day_delay` with variable `delay_rate` to calculate delay rate by day of month
day_delay <- years %>%
select(DayofMonth, status) %>%
group_by(DayofMonth) %>%
summarize(num_delays = sum(status == 'Delayed'),
num_flights = n(),
delay_rate = sum(status == 'Delayed') / n())
# Line plot of **delay rate by day of month**
plot_day <- day_delay %>%
ggplot(aes(x = DayofMonth, y = delay_rate, group = 1)) +
geom_line(color = "steelblue") +
geom_point(color = "steelblue") +
labs(title = "Day of Month", y = "Delay Rate") +
theme_classic() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
plot_day
Delay rate by day of week (Line plot)
# Creating plot labels for `DayOfWeek`
week_label <- c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun")
# Creating data frame `day_delay` with variable `delay_rate` to calculate delay rate by day of week
week_delay <- years %>%
select(DayOfWeek, status) %>%
group_by(DayOfWeek) %>%
summarize(num_delays = sum(status == 'Delayed'),
num_flights = n(),
delay_rate = sum(status == 'Delayed') / n())
# Line plot of **delay rate by day of week**
plot_week <- week_delay %>%
ggplot(aes(x = DayOfWeek, y = delay_rate, group = 1)) +
geom_line(color = "steelblue") +
geom_point(color = "steelblue") +
scale_x_discrete(labels = week_label) +
labs(title = "Day of Week", y = "Delay Rate") +
theme_classic() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
plot_week
Preparing data for delay rate by hour (Part 1) – Plot is for own reference
# Creating data frame `all_hours` to categorise all `DepTime` as "Normal" or "Unusual"
all_hours <- years %>%
select(DepTime, status) %>%
drop_na(DepTime) %>%
mutate(dep_hour = ifelse(
nchar(DepTime) == 1 | nchar(DepTime) == 2 | DepTime > 2400,
"Unusual", "Normal"))
# Calculating % of "Unusual" DepTime
perc_hour <- all_hours %>%
count(dep_hour) %>%
mutate(hour_perc = n / nrow(all_hours) * 100)
# Creating plot labels for `hour_perc`
perc_label1 <- c("0", "25", "50", "75", "100")
# Bar plot of % of Normal vs Unusual timings
perc_hour %>%
ggplot(aes(x = dep_hour, y = hour_perc)) +
geom_bar(stat = "identity", fill = "steelblue") +
scale_y_continuous(labels = paste0(perc_label1, "%")) +
geom_text(aes(label = paste0(round(hour_perc,2), "%")), position = position_dodge(width = 0.9), vjust = -0.4) +
labs(title = "Percentage of Normal vs Unusual Departure Timings", x = "", y = "Percentage") +
theme_bw() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12)) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
Since only 0.33% of the departure times from variable
DepTime is categorised as “Unusual”, omitting this data
will neither skew nor affect the result.
Preparing data for delay rate by hour (Part 2)
# Logical Data cleansing
time_delay <- years %>%
select(DepTime, status) %>%
filter(nchar(DepTime) == 3) %>%
mutate(DepTime = signif(DepTime, 1)) %>%
mutate(DepTime = case_when(
DepTime == 100 ~ "0100",
DepTime == 200 ~ "0200",
DepTime == 300 ~ "0300",
DepTime == 400 ~ "0400",
DepTime == 500 ~ "0500",
DepTime == 600 ~ "0600",
DepTime == 700 ~ "0700",
DepTime == 800 ~ "0800",
DepTime == 900 ~ "0900",
TRUE ~ "1000"))
time_delay$DepTime <- factor(time_delay$DepTime)
time_delay2 <- years %>%
select(DepTime, status) %>%
filter(nchar(DepTime) == 4) %>%
mutate(DepTime = signif(DepTime, 2)) %>%
mutate(DepTime = case_when(
DepTime > 2400 ~ paste0("0",as.character(DepTime - 2400)),
TRUE ~ as.character(DepTime)))
time_delay2$DepTime <- factor(time_delay2$DepTime)
time_delay3 <- years %>%
select(DepTime, status) %>%
filter(nchar(DepTime) == 1 |
nchar(DepTime) == 2) %>%
mutate(DepTime = case_when(
nchar(DepTime) == 1 ~ "2400",
nchar(DepTime) == 2 ~ "2400",
TRUE ~ as.character(DepTime)))
time_delay3$DepTime <- factor(time_delay3$DepTime)
# Data frame of cleaned data `normal_hours`, with `DepTime` factored by hour in `new_hour`
time_delay <- rbind(time_delay, time_delay2, time_delay3)
summary(time_delay$DepTime)## 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000
## 19762 7746 1996 2116 124711 1314650 1368275 1445383 1350730 1357816
## 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
## 1332094 1357413 1322576 1304779 1280013 1323622 1363610 1321358 1224149 984434
## 2100 2200 2300 2400
## 801662 418428 188940 103058Delay rate by hour (Line plot)
# Creating data frame `hour_delay` with variable `delay_rate` to calculate delay rate by hour
time_delay <- time_delay %>%
group_by(DepTime) %>%
summarize(num_delays = sum(status == 'Delayed'),
num_flights = n(),
delay_rate = num_delays / num_flights)
# Line plot of **delay rate by hour**
plot_hour <- time_delay %>%
ggplot(aes(x = DepTime, y = delay_rate, group = 1)) +
geom_line(color = "steelblue") +
geom_point(color = "steelblue") +
labs(title = "Hour", y = "Delay Rate") +
theme_classic() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
plot_hour
Grid of all plots using cowplot package
p <- plot_grid(plot_month, plot_day, plot_week, plot_hour, scale = 1)
title <- ggdraw() + draw_label("Delay Rates Based On", fontface='bold', size = 17)
plot_grid(title, p, ncol=1, rel_heights=c(0.1, 1))
# Removing data frames from Question 1
rm(day_delay, month_delay, time_delay, week_delay)2. Do older planes suffer more delays?
Preparation of data frame carrier
# Creating data frame `carrier`
carrier <- years %>%
select(UniqueCarrier, TailNum, DepDelay, ArrDelay, Year, CarrierDelay, WeatherDelay, NASDelay, SecurityDelay, LateAircraftDelay)
colnames(carrier)[2] <- "tailnum"
# Removing NA values from `planes` dataset and naming it `plane_year`
planes$year <- as.numeric(planes$year)## Warning: NAs introduced by coercionplanes <- subset(planes, is.na(year) == F)
summary(planes$year)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0 1992 1999 1996 2002 2008plane_year <- planes %>%
select(tailnum, year) %>%
filter(year != 0)
# Joining `plane_year` and `carrier` using inner join
colnames(plane_year)[2] <- "planeyear"
carrier <- carrier %>%
inner_join(plane_year) ## joining by tailnum## Joining, by = "tailnum"carrier$Year <- as.integer(as.character(carrier$Year))Summary of data frame carrier
summary(carrier)## UniqueCarrier tailnum DepDelay ArrDelay
## Length:16633761 Length:16633761 Min. :-1200.00 Min. :-692.00
## Class :character Class :character 1st Qu.: -4.00 1st Qu.: -9.00
## Mode :character Mode :character Median : 0.00 Median : -1.00
## Mean : 10.16 Mean : 9.05
## 3rd Qu.: 9.00 3rd Qu.: 13.00
## Max. : 2601.00 Max. :2598.00
## NA's :109765 NA's :147012
## Year CarrierDelay WeatherDelay NASDelay
## Min. :2005 Min. : 0.000 Min. : 0.0000 Min. : -49.000
## 1st Qu.:2005 1st Qu.: 0.000 1st Qu.: 0.0000 1st Qu.: 0.000
## Median :2006 Median : 0.000 Median : 0.0000 Median : 0.000
## Mean :2006 Mean : 3.558 Mean : 0.7186 Mean : 3.806
## 3rd Qu.:2007 3rd Qu.: 0.000 3rd Qu.: 0.0000 3rd Qu.: 0.000
## Max. :2007 Max. :2580.000 Max. :1510.0000 Max. :1392.000
##
## SecurityDelay LateAircraftDelay planeyear
## Min. : 0.0000 Min. : 0.00 Min. :1956
## 1st Qu.: 0.0000 1st Qu.: 0.00 1st Qu.:1992
## Median : 0.0000 Median : 0.00 Median :1999
## Mean : 0.0255 Mean : 4.53 Mean :1997
## 3rd Qu.: 0.0000 3rd Qu.: 0.00 3rd Qu.:2002
## Max. :366.0000 Max. :1366.00 Max. :2007
## str(carrier)## 'data.frame': 16633761 obs. of 11 variables:
## $ UniqueCarrier : chr "UA" "UA" "UA" "UA" ...
## $ tailnum : chr "N935UA" "N941UA" "N342UA" "N326UA" ...
## $ DepDelay : int -2 -6 -7 -9 34 102 106 -1 -1 17 ...
## $ ArrDelay : int -18 -23 -24 -33 3 97 85 12 -18 17 ...
## $ Year : int 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005 ...
## $ CarrierDelay : int 0 0 0 0 0 23 46 0 0 17 ...
## $ WeatherDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ NASDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ SecurityDelay : int 0 0 0 0 0 0 0 0 0 0 ...
## $ LateAircraftDelay: int 0 0 0 0 0 74 39 0 0 0 ...
## $ planeyear : num 1992 1992 1988 1988 1990 ...Histogram of year of manufacture of planes from
plane_year
hist <- plane_year %>%
ggplot(aes(x=planeyear)) +
geom_histogram(fill = "steelblue", binwidth = 1) +
labs(title = "Year of Manufacture", y = "Frequency") +
geom_vline(xintercept=c(1980), linetype="dotted") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(axis.title.x = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold")) +
theme(legend.position = "none") Line charts for mean departure and arrival delays
# Preparing data for line charts
plane_age <- carrier %>%
mutate(age = Year - planeyear) %>%
group_by(age) %>%
summarise(
mean_depdelay = mean(DepDelay, na.rm = TRUE),
mean_arrdelay = mean(ArrDelay, na.rm = TRUE))
summary(plane_age$age)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -2.00 11.25 24.50 24.50 37.75 51.00plane_age <- plane_age %>%
filter(age > 0) # To remove negative values# Line chart of Mean departure delay (all planes)
p1 <- ggplot(plane_age, aes(x=age, y=mean_depdelay, group=1)) +
geom_line(color = "blue") +
labs(x = "Years of service", y = "Mean Departure Delay (in minutes)") +
geom_vline(xintercept=c(25), linetype="dotted") +
theme_classic()
# Line charts of Mean arrival delay (all planes)
p2 <- ggplot(plane_age, aes(x=age, y=mean_arrdelay, group=1)) +
geom_line(color = "blue") +
labs(x = "Years of service", y = "Mean Arrival Delay (in minutes)") +
geom_vline(xintercept=c(25), linetype="dotted") +
theme_classic()
# Grid of all plots using `cowplot` package
bottom_row1 <- plot_grid(p1,p2)
plot_grid(hist, bottom_row1, nrow = 2)
Line charts of mean departure and arrival delays based on year of make
# Pre 1980 (Older) Line chart
older_planes <- plane_age %>%
filter(age > 25)
colnames(older_planes) <- c("age", "Mean Departure Delay", "Mean Arrival Delay")
older_planes <- melt(older_planes, id.vars = "age")
plot_cols = c("#e48f1b", "#619ed6")
p3 <- ggplot(older_planes, aes(x = age, y = value)) +
geom_line(aes(color = variable)) +
labs(title = "Planes manufactured before the 1980s" ,x = "Years of service", y = "Mean Delay (in minutes)") +
scale_color_manual(values = plot_cols) +
ylim(2,15) +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p3
# Post 1980 (Younger) Line chart
younger_planes <- plane_age %>%
filter(age < 25)
colnames(younger_planes) <- c("age", "Mean Departure Delay", "Mean Arrival Delay")
younger_planes <- melt(younger_planes, id.vars = "age")
p4 <- ggplot(younger_planes, aes(x = age, y = value)) +
geom_line(aes(color = variable)) +
labs(title = "Planes manufactured after the 1980s" ,x = "Years of service", y = "Mean Delay (in minutes)") +
scale_color_manual(values = plot_cols) +
ylim(2,15) +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p4
# Grid plot of both line charts
plot_grid(p3,p4)
Grouped bar chart of delay factors
# Creating new data frame with mean values of all delay factors
delay_factors <- carrier %>%
mutate(age = Year - planeyear) %>%
filter(age > 0) %>%
group_by(age) %>%
summarise(
carrier_delay = mean(CarrierDelay, na.rm = TRUE),
weather_delay = mean(WeatherDelay, na.rm = TRUE),
nas_delay = mean(NASDelay, na.rm = TRUE),
security_delay = mean(SecurityDelay, na.rm = TRUE),
aircraft_delay = mean(LateAircraftDelay, na.rm = TRUE))
colnames(delay_factors) <- c("age", "Carrier Delay", "Weather Delay", "NAS Delay", "Security Delay", "Late Aircraft Delay")
# Data preparation
delay_factors <- melt(delay_factors, id.vars = "age")
delay_factors <- delay_factors %>%
mutate(group = ifelse(
age > 25, "Over 25 Years", "Below 25 Years"))
delay_factors <- dcast(delay_factors, group + variable ~ ., sum)
colnames(delay_factors)[3] <- "value"
grouped_bar <- delay_factors %>%
ggplot(aes(fill=group, y=value, x=variable)) +
geom_bar(position="dodge", stat = "identity", width = 0.7) +
scale_y_continuous(labels = scales::comma) +
scale_color_manual(values = plot_cols) +
labs(title = "Mean Delay based on Delay Factors ", x = "", y = "Mean Delay (minutes)") +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
grouped_bar
3 by 1 Line chart (Top 3 delay factors)
# Creating new data frame with mean values of the top 3 delay factors
delay_lines <- carrier %>%
mutate(age = Year - planeyear) %>%
filter(age > 0) %>%
group_by(age) %>%
summarise(
mean_carrierdelay = mean(CarrierDelay, na.rm = TRUE),
mean_NASdelay = mean(NASDelay, na.rm = TRUE),
mean_aircraftdelay = mean(LateAircraftDelay, na.rm = TRUE))
# 3x1 Line chart
p5 <- ggplot(delay_lines, aes(x=age, y=mean_carrierdelay, group=1)) +
geom_line(color = "blue") +
ylim(0,7) +
labs(x = "Years of service", y = "Mean Carrier Delay (in minutes)") +
theme_classic()
p5
p6 <- ggplot(delay_lines, aes(x=age, y=mean_NASdelay, group=1)) +
geom_line(color = "blue") +
ylim(0,7) +
labs(x = "Years of service", y = "Mean NAS Delay (in minutes)") +
theme_classic()
p6
p7 <- ggplot(delay_lines, aes(x=age, y=mean_aircraftdelay, group=1)) +
geom_line(color = "blue") +
ylim(0,7) +
labs(x = "Years of service", y = "Mean Late Aircraft Delay (in minutes)") +
theme_classic()
p7
# Grid plot of grouped bar chart + 3x1 Line chart
bottom_row2 <- plot_grid(p5, p6, p7, nrow = 1)
plot_grid(grouped_bar, bottom_row2, ncol = 1)
# Removing data frames from Question 2
rm(carrier, plane_age, younger_planes, delay_factors, delay_lines)3. How does the number of people flying between different locations change over time?
Creating Origin-Destination pairs in data frame
od_pairs
# Origin-Dest pairs
od_pairs <- years %>%
group_by(Year, Origin, Dest) %>%
summarize(num_trips = n()) %>%
arrange(desc(num_trips))## `summarise()` has grouped output by 'Year', 'Origin'. You can override using
## the `.groups` argument.od_pairs$combi <- paste0(od_pairs$Origin,"/",od_pairs$Dest)
od_pairs$combi <- as.factor(od_pairs$combi)
od_pairs$Year <- as.factor(od_pairs$Year)
# New data frame `sum_od` based on total number of trips in each combination
sum_od <- od_pairs %>%
group_by(combi) %>%
mutate(sum_trips = sum(num_trips)) %>%
arrange(desc(sum_trips))
sum_od <- head(sum_od, n = 30) # finding the top 10 OD in 3 years(For own reference) O-D pairs with most traffic in 3 years
# Bar chart
sum_od %>%
ggplot(aes(y = combi, x = sum_trips)) +
geom_bar(stat = "identity", width = 0.7, fill = "steelblue") +
scale_x_continuous(labels = comma) +
labs(title = "Origin-Destination Combinations with most trips", x = "Count", y = "Origin/Destination") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(legend.position = "right") 
Finding top O-D pairs
# Filtering data frame to the top 10 OD combinations
top_od_pairs <- od_pairs %>%
filter(combi == "SAN/LAX" |
combi == "OGG/HNL" |
combi == "LGA/DCA" |
combi == "LGA/BOS" |
combi == "LAX/SAN" |
combi == "LAX/LAS" |
combi == "LAS/LAX" |
combi == "HNL/OGG" |
combi == "DCA/LGA" |
combi == "BOS/LGA")
plot_cols = c("#e48f1b", "steelblue", "#aeaeae")
# Stacked bar chart of top 10 ODs
p1 <- top_od_pairs %>%
ggplot(aes(fill = Year, y = combi, x = num_trips)) +
geom_bar(position = "stack", stat = "identity", width = 0.7) +
scale_x_continuous(labels = comma) +
scale_fill_manual(values = plot_cols) +
labs(title = "Top 10 Origin-Destination Combinations", x = "Number of trips", y = "Origin/Destination") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p1
Data preparation for ODs with significant changes
# Manipulating of dataset
unusual_pairs <- od_pairs %>%
filter(Year == 2005) %>%
mutate(num_2005 = num_trips)
unusual_pairs = subset(unusual_pairs, select = c("combi", "num_2005"))
unusual_pairs1 <- od_pairs %>%
filter(Year == 2006) %>%
mutate(num_2006 = num_trips)
unusual_pairs1 = subset(unusual_pairs1, select = c("combi", "num_2006"))
unusual_pairs2 <- od_pairs %>%
filter(Year == 2007) %>%
mutate(num_2007 = num_trips)
unusual_pairs2 = subset(unusual_pairs2, select = c("combi", "num_2007"))
# Using inner join to combine all 3 datasets and retrieving top 10 values
odd_pairs <- unusual_pairs %>%
inner_join(unusual_pairs1) %>%
inner_join(unusual_pairs2) ## Joining, by = "combi"
## Joining, by = "combi"odd_pairs$max <- pmax(odd_pairs$num_2005, odd_pairs$num_2006, odd_pairs$num_2007)
odd_pairs$min <- pmin(odd_pairs$num_2005, odd_pairs$num_2006, odd_pairs$num_2007)
odd_pairs$max_diff <- odd_pairs$max - odd_pairs$min
odd_pairs1 <- odd_pairs %>%
arrange(desc(max_diff))
odd_pairs1 <- head(odd_pairs1, n = 10)
# More data manipulation
odd_pairs1 <- melt(odd_pairs1, id.vars = "combi")
odd_pairs1 <- odd_pairs1 %>%
mutate(case_when(
variable == "num_2005" ~ "2005",
variable == "num_2006" ~ "2006",
variable == "num_2007" ~ "2007"))
colnames(odd_pairs1) <- c("combi", "variable", "value", "Year")
odd_pairs1 <- odd_pairs1 %>%
filter(variable != "max_diff")
odd_pairs1 <- head(odd_pairs1, n = 30)Plot of top 10 OD combinations with significant changes
# Stacked bar chart
p2 <- odd_pairs1 %>%
ggplot(aes(fill = Year, y = combi, x = value)) +
geom_bar(position = "stack", stat = "identity", width = 0.7) +
scale_x_continuous(labels = comma) +
scale_fill_manual(values = plot_cols) +
labs(title = "Origin-Destination Combinations with significant changes in traffic", x = "Number of trips", y = "Origin/Destination") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p2
Grid plot of both plot using cowplot package
plot_grid(p1,p2)
Line chart of top 5 combinations
line_pairs <- years %>%
select(Year, Month, DayofMonth, Origin, Dest) %>%
group_by(Year, Month, DayofMonth, Origin, Dest) %>%
summarize(num_trips = n()) %>%
arrange(desc(num_trips))## `summarise()` has grouped output by 'Year', 'Month', 'DayofMonth', 'Origin'.
## You can override using the `.groups` argument.line_pairs$combi <- paste0(line_pairs$Origin,"/",line_pairs$Dest)
line_pairs$combi <- as.factor(line_pairs$combi)
line_pairs$Year <- as.factor(line_pairs$Year)
top_5 <- line_pairs %>%
filter(combi == "SAN/LAX" |
combi == "LAX/SAN" |
combi == "LAX/LAS" |
combi == "LAS/LAX" |
combi == "BOS/LGA")
top_5$date <- as.Date(paste0(top_5$Year,"-",top_5$Month,"-",top_5$DayofMonth))
top_5 <- top_5 %>%
group_by(month = floor_date(date, unit = "month"))
mean_line_pairs <- top_5 %>%
group_by(combi, month) %>%
summarise(
mean_trips = mean(num_trips)
)## `summarise()` has grouped output by 'combi'. You can override using the
## `.groups` argument.plot_cols2 = c("steelblue", "#e48f1b", "#aeaeae", "#f7d027", "steelblue2")
p3 <- ggplot(mean_line_pairs, aes(x = month, y = mean_trips, color = combi)) +
geom_line(size = 0.5) +
scale_color_manual(values = plot_cols2) +
labs(title = "Traffic of Top 5 Origin-Destination Combinations over 3 years" , x = "", y = "Mean number of trips") +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p3
Line chart of top 5 combinations with the most changes
change_5 <- line_pairs %>%
filter(combi == "OGG/HNL" |
combi == "HNL/OGG" |
combi == "KOA/HNL" |
combi == "HNL/LIH" |
combi == "HNL/KOA")
change_5$date <- as.Date(paste0(change_5$Year,"-",change_5$Month,"-",change_5$DayofMonth))
change_5 <- change_5 %>%
group_by(month = floor_date(date, unit = "month"))
mean_change_pairs <- change_5 %>%
group_by(combi, month) %>%
summarise(
mean_trips = mean(num_trips))## `summarise()` has grouped output by 'combi'. You can override using the
## `.groups` argument.p4 <- ggplot(mean_change_pairs, aes(x = month, y = mean_trips, color = combi)) +
geom_line(size = 0.5) +
scale_color_manual(values = plot_cols2) +
labs(title = "Traffic of Origin-Destination Combinations with significant increases over 3 years" , x = "", y = "Mean number of trips") +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p4
Grid plot of both plot using cowplot package
plot_grid(p3,p4)
# Removing data frames from Question 3
rm(od_pairs, sum_od, top_od_pairs, unusual_pairs, unusual_pairs1, unusal_pairs2, odd_pairs, odd_pairs1, line_pairs, top5)## Warning in rm(od_pairs, sum_od, top_od_pairs, unusual_pairs, unusual_pairs1, :
## object 'unusal_pairs2' not found## Warning in rm(od_pairs, sum_od, top_od_pairs, unusual_pairs, unusual_pairs1, :
## object 'top5' not found4. Can you detect cascading failures as delays in one airport create delays in others?
Preparing data frame
# Setting variables as factors
years$Month <- as.factor(years$Month)
years$DayofMonth <- as.factor(years$DayofMonth)
# Adding new column to data for flight status
years$dep_delayed <- "No"
years$dep_delayed[years$DepDelay > 0] <- "Yes"
years$arr_delayed <- "No"
years$arr_delayed[years$ArrDelay > 0] <- "Yes"
years$dep_delayed <- as.factor(years$dep_delayed)
years$arr_delayed <- as.factor(years$arr_delayed)
summary(years$dep_delayed)## No Yes
## 13227403 8508330summary(years$arr_delayed)## No Yes
## 12021690 9714043Delay rate per month (From question 1)
# plot for delay rate per month (from question 1)
month_label <- c("Jan","Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov","Dec")
month_delay <- years %>%
select(Month, dep_delayed, arr_delayed) %>%
group_by(Month) %>%
summarise(num_delays = sum(dep_delayed == 'Yes' | arr_delayed == 'Yes'),
num_flights = n(),
delay_rate = num_delays / n())
p1 <- month_delay %>%
ggplot(aes(x = Month, y = delay_rate, group = 1)) +
geom_line(color = "steelblue") +
geom_point(color = "steelblue") +
scale_x_discrete(labels = month_label) +
labs(title = "Delay Rate per Month", y = "Delay Rate (in minutes)") +
theme_classic() +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
p1
Histogram for number of flights by month
bar_grouped <- years %>%
select(Month, Year) %>%
group_by(Month, Year) %>%
summarise(num_flights = n())## `summarise()` has grouped output by 'Month'. You can override using the
## `.groups` argument.bar_grouped$Year <- factor(bar_grouped$Year)
plot_cols = c("#e48f1b", "steelblue", "#aeaeae")
p2 <- bar_grouped %>%
ggplot(aes(fill = Year, y = num_flights, x = Month)) +
geom_bar(position="stack", stat = "identity", width = 0.7) +
scale_fill_manual(values = plot_cols) +
scale_x_discrete(labels = month_label) +
scale_y_continuous(labels = scales::comma) +
labs(title = "Number of flights", x = "", y = "Frequency") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
# Grid plot using `cowplot` package
plot_grid(p1,p2, align = "h")
Third plot: Departure and Arrival rates in December (Month with highest delay rate)
first_month <- years %>%
select(Month, dep_delayed, arr_delayed, DayofMonth) %>%
filter(Month == 12) %>%
group_by(DayofMonth) %>%
summarise(num_arr = sum(arr_delayed == 'Yes'),
num_flights = n(),
arr_rate = num_arr / n(),
num_dep = sum(dep_delayed == 'Yes'),
num_flights = n(),
dep_rate = num_dep / n())
first_month <- first_month[c(1,4,6)]
colnames(first_month) <- c("DayofMonth", "Arrival Delay Rate", "Departure Delay Rate")
first_month <- melt(first_month, id.vars = "DayofMonth")
str(first_month)## 'data.frame': 62 obs. of 3 variables:
## $ DayofMonth: Factor w/ 31 levels "1","2","3","4",..: 1 2 3 4 5 6 7 8 9 10 ...
## $ variable : Factor w/ 2 levels "Arrival Delay Rate",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ value : num 0.48 0.488 0.432 0.479 0.454 ...p3 <- ggplot(first_month, aes(x = DayofMonth, y = value, group = variable)) +
geom_line(aes(color = variable)) +
labs(title = "Delay Rate in December" , x = "", y = "") +
scale_color_manual(values = plot_cols) +
theme_bw() +
theme(legend.title=element_blank()) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p3
Preparing data for fourth plot: Delay rate on 22nd December (Day with highest delay)
# Logical data cleaning
day_aug <- years %>%
select(Month, DayofMonth, DepTime, CRSDepTime, DepDelay, dep_delayed, arr_delayed) %>%
filter(Month == 12,
DayofMonth == 22 |
DayofMonth == 23) %>%
drop_na(DepTime) %>%
filter(nchar(DepTime) > 2)
day_aug1 <- day_aug %>%
filter(nchar(DepTime) == 3) %>%
mutate(DepTime = signif(DepTime, 1)) %>%
mutate(DepTime = case_when(
DepTime == 100 ~ "0100",
DepTime == 200 ~ "0200",
DepTime == 300 ~ "0300",
DepTime == 400 ~ "0400",
DepTime == 500 ~ "0500",
DepTime == 600 ~ "0600",
DepTime == 700 ~ "0700",
DepTime == 800 ~ "0800",
DepTime == 900 ~ "0900",
TRUE ~ "1000"))
day_aug2 <- day_aug %>%
filter(nchar(DepTime) == 4) %>%
mutate(DepTime = signif(DepTime, 2)) %>%
mutate(DepTime = case_when(
DepTime == 2400 ~ "2400",
DepTime > 2400 ~ paste0("0",as.character(DepTime - 2400)),
TRUE ~ as.character(DepTime)))
day_aug <- rbind(day_aug1, day_aug2)
day_aug$DepTime <- factor(day_aug$DepTime)
summary(day_aug$DepTime)## 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600
## 245 98 31 18 543 6594 7144 7454 6971 6898 7034 7066 6739 6944 6794 6998
## 1700 1800 1900 2000 2100 2200 2300 2400
## 7159 6980 6445 5518 4567 2923 1544 223day_delay <- day_aug %>%
select(DayofMonth, DepTime, dep_delayed, arr_delayed) %>%
group_by(DepTime, DayofMonth) %>%
summarise(num_arr = sum(arr_delayed == 'Yes'),
num_flights = n(),
arr_rate = num_arr / n(),
num_dep = sum(dep_delayed == 'Yes'),
num_flights = n(),
dep_rate = num_dep / n())## `summarise()` has grouped output by 'DepTime'. You can override using the
## `.groups` argument.day_delay$DepTime <- as.character(day_delay$DepTime)
day_delay <- day_delay[c(1,2,5,7)]
colnames(day_delay) <- c("DepTime", "DayofMonth", "Arrival Delay Rate", "Departure Delay Rate")
day_delay$DepTime <- factor(day_delay$DepTime)
day_delay <- melt(day_delay, id.vars = c("DepTime", "DayofMonth"))
str(day_delay)## 'data.frame': 96 obs. of 4 variables:
## $ DepTime : Factor w/ 24 levels "0100","0200",..: 1 1 2 2 3 3 4 4 5 5 ...
## $ DayofMonth: Factor w/ 31 levels "1","2","3","4",..: 22 23 22 23 22 23 22 23 22 23 ...
## $ variable : Factor w/ 2 levels "Arrival Delay Rate",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ value : num 0.909 0.926 0.861 0.919 0.833 ...Fourth plot
p4 <- day_delay %>%
filter(DayofMonth == 22) %>%
arrange(DepTime) %>%
ggplot(aes(x = DepTime, y = value, group = variable)) +
geom_line(aes(color = variable)) +
labs(title = "Delay Rate on 22nd December" , x = "", y = "") +
scale_color_manual(values = plot_cols) +
theme_bw() +
theme(legend.title=element_blank()) +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank()) +
theme(plot.title = element_text(hjust = 0.5,size = 13, face = "bold"))
p4
Grid plot using cowplot package
plot_grid(p3, p4, ncol=1)
Manually checking for cascading failures due to delays
check <- years %>%
select(Year, Month, DayofMonth, CRSDepTime, DepTime, CRSArrTime, ArrTime, TailNum, FlightNum, Origin, Dest, arr_delayed, dep_delayed) %>%
filter(Year == 2005,
Month == 12,
DayofMonth == 22,
arr_delayed == "Yes",
dep_delayed == "Yes",
DepTime > 1100,
DepTime < 1300)
# Found 2 examples of 1st degree cascading failures due to delays
check <- check %>%
filter(TailNum == "N957SW" |
TailNum == "N835AE")
summary(check)## Year Month DayofMonth CRSDepTime DepTime CRSArrTime
## 2005:4 12 :4 22 :4 Min. :1010 Min. :1105 Min. :1116
## 2006:0 1 :0 1 :0 1st Qu.:1014 1st Qu.:1105 1st Qu.:1119
## 2007:0 2 :0 2 :0 Median :1080 Median :1172 Median :1268
## 3 :0 3 :0 Mean :1082 Mean :1176 Mean :1276
## 4 :0 4 :0 3rd Qu.:1148 3rd Qu.:1242 3rd Qu.:1426
## 5 :0 5 :0 Max. :1155 Max. :1253 Max. :1454
## (Other):0 (Other):0
## ArrTime TailNum FlightNum Origin
## Min. :1213 Length:4 Min. :4688 Length:4
## 1st Qu.:1215 Class :character 1st Qu.:4688 Class :character
## Median :1332 Mode :character Median :5559 Mode :character
## Mean :1357 Mean :5612
## 3rd Qu.:1474 3rd Qu.:6483
## Max. :1550 Max. :6642
##
## Dest arr_delayed dep_delayed
## Length:4 No :0 No :0
## Class :character Yes:4 Yes:4
## Mode :character
##
##
##
## # Removing data frames from Question 4
rm(month_delay, bar_grouped, first_month, day_aug, day_aug1, day_delay, check)5. Use the available variables to construct a model that predicts delay.
I will be building a model to predict DepDelay, or
departure delay.
Preparing data for exploratory data analysis
# Checking missing values (missing values or empty values)
colSums(is.na(years)|years=='') ## Year Month DayofMonth DayOfWeek
## 0 0 0 0
## DepTime CRSDepTime ArrTime CRSArrTime
## 416412 0 463805 0
## UniqueCarrier FlightNum TailNum ActualElapsedTime
## 0 0 22 463805
## CRSElapsedTime AirTime ArrDelay DepDelay
## 998 463805 463805 416412
## Origin Dest Distance TaxiIn
## 0 0 0 0
## TaxiOut Cancelled CancellationCode Diverted
## 0 0 21319319 0
## CarrierDelay WeatherDelay NASDelay SecurityDelay
## 0 0 0 0
## LateAircraftDelay status dep_delayed arr_delayed
## 0 0 0 0# Creating new feature for status of flights
years$status <- NA
years$status[years$Diverted == 1] <- "Diverted"
years$status[years$DepDelay > 0] <- "Delayed"
years$status[years$Cancelled == 1] <- "Cancelled"
years$status <- ifelse(years$Diverted != 1 & years$DepDelay <= 0 &
years$Cancelled != 1, "On Time", years$status)
years$status <- as.factor(years$status)Factors 1 and 2: CRSDepTime and
CRSArrTime
summary(years$CRSDepTime) ## Based on data frame, timings with "0" refers to 24:00, or 00:00## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0 930 1325 1334 1721 2359summary(years$CRSArrTime) ## Based on data frame, timings with "0" refers to 24:00, or 00:00## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0 1116 1520 1497 1909 2400main <- years %>% ## Included a few rows for convenience -- For machine learning section
select(TailNum, Year, CRSDepTime, CRSArrTime, DepDelay, CRSElapsedTime, status) %>%
drop_na(DepDelay) ## NA values are because flight is cancelled
summary(main) ## TailNum Year CRSDepTime CRSArrTime
## Length:21319321 2005:7006866 Min. : 0 Min. : 0
## Class :character 2006:7019988 1st Qu.: 930 1st Qu.:1115
## Mode :character 2007:7292467 Median :1325 Median :1520
## Mean :1333 Mean :1496
## 3rd Qu.:1720 3rd Qu.:1908
## Max. :2359 Max. :2400
##
## DepDelay CRSElapsedTime status
## Min. :-1200.00 Min. : -97.0 Cancelled: 0
## 1st Qu.: -4.00 1st Qu.: 76.0 Delayed : 8508330
## Median : 0.00 Median : 109.0 Diverted : 21996
## Mean : 10.07 Mean : 127.1 On Time :12788995
## 3rd Qu.: 9.00 3rd Qu.: 157.0
## Max. : 2601.00 Max. :1031.0
## NA's :727## Analysis:
## Single digit CRSDepTime represents the minutes in the hour 2400, eg CRSDepTime == 8 = 2408; same thing for DepTime. Timings over 2400 indicates next day -- hence take (Time - 2400)Data cleaning for CRSDepTime and
CRSArrTime
# Rounding off values to nearest hour and making it a factor from **0100 to 1000** (3 digit timings)
crs1 <- main %>%
filter(nchar(CRSArrTime) == 3) %>%
mutate(CRSArrTime = signif(CRSArrTime, 1)) %>%
mutate(CRSArrTime = case_when(
CRSArrTime == 100 ~ "0100",
CRSArrTime == 200 ~ "0200",
CRSArrTime == 300 ~ "0300",
CRSArrTime == 400 ~ "0400",
CRSArrTime == 500 ~ "0500",
CRSArrTime == 600 ~ "0600",
CRSArrTime == 700 ~ "0700",
CRSArrTime == 800 ~ "0800",
CRSArrTime == 900 ~ "0900",
TRUE ~ "1000"))
crs1$CRSArrTime <- factor(crs1$CRSArrTime)
crs2 <- main %>%
filter(nchar(CRSDepTime) == 3) %>%
mutate(CRSDepTime = signif(CRSDepTime, 1)) %>%
mutate(CRSDepTime = case_when(
CRSDepTime == 100 ~ "0100",
CRSDepTime == 200 ~ "0200",
CRSDepTime == 300 ~ "0300",
CRSDepTime == 400 ~ "0400",
CRSDepTime == 500 ~ "0500",
CRSDepTime == 600 ~ "0600",
CRSDepTime == 700 ~ "0700",
CRSDepTime == 800 ~ "0800",
CRSDepTime == 900 ~ "0900",
TRUE ~ "1000"))
crs2$CRSDepTime <- factor(crs2$CRSDepTime)
# Rounding off values to nearest hour and making it a factor from **1000 to 2400** (4 digit timings)
crs3 <- main %>%
filter(nchar(CRSDepTime) == 4) %>%
arrange(CRSDepTime) %>%
mutate(CRSDepTime = signif(CRSDepTime, 2))
crs4 <- main %>%
filter(nchar(CRSArrTime) == 4) %>%
arrange(CRSArrTime) %>%
mutate(CRSArrTime = signif(CRSArrTime, 2))
crs3$CRSDepTime <- factor(crs3$CRSDepTime)
crs4$CRSArrTime <- factor(crs4$CRSArrTime)
# Cleaning of single digit timings and rounding it to **2400*
crs5 <- main %>%
filter(nchar(CRSDepTime) == 1 |
nchar(CRSDepTime) == 2) %>%
mutate(CRSDepTime = case_when(
nchar(CRSDepTime) == 1 ~ "2400",
nchar(CRSDepTime) == 2 ~ "2400",
TRUE ~ as.character(CRSDepTime)))
crs6 <- main %>%
filter(nchar(CRSArrTime) == 1 |
nchar(CRSArrTime) == 2) %>%
mutate(CRSArrTime = case_when(
nchar(CRSArrTime) == 1 ~ "2400",
nchar(CRSArrTime) == 2 ~ "2400",
TRUE ~ as.character(CRSArrTime)))
crs5$CRSDepTime <- factor(crs5$CRSDepTime)
crs6$CRSArrTime <- factor(crs6$CRSArrTime)
# Binding of all dataframes together
arr <- rbind(crs1, crs4, crs6)
dep <- rbind(crs2, crs3, crs5)
rm(crs1, crs2, crs3, crs4, crs5, crs6)Plot of CRSDepTime and CRSArrTime against
mean departure delay rates
# CRSDepTime against mean departure delay rates
dep1 <- dep %>%
group_by(CRSDepTime) %>%
summarise(mean_delay = mean(DepDelay, na.rm = TRUE))
# CRSArrTime against mean departure delay rates
arr1 <- arr %>%
group_by(CRSArrTime) %>%
summarise(mean_delay = mean(DepDelay, na.rm = TRUE)) %>%
arrange(CRSArrTime)
p1 <- ggplot(dep1, aes(x = CRSDepTime, y = mean_delay, group = 1)) +
geom_line(color = "steelblue", size = 0.7) +
geom_point(color = "steelblue") +
labs(title = "Scheduled Departure Time" , x = "", y = "Mean Delay (minutes)") +
ylim(0,20) +
theme_classic() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
p2 <- ggplot(arr1, aes(x = CRSArrTime, y = mean_delay, group = 1)) +
geom_line(color = "steelblue", size = 0.7) +
geom_point(color = "steelblue") +
labs(title = "Scheduled Arrival Time" , x = "", y = "Mean Delay (minutes)") +
ylim(0,20) +
theme_classic() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12),
axis.title.x = element_blank())
title <- ggdraw() + draw_label("Mean Departure Delay based on",
fontface='bold', size = 17)
p <- plot_grid(p1,p2)
plot_grid(title, p, ncol=1, rel_heights=c(0.1, 1))
Factor 3: age of Planes
(See question 2 plot) I will be including age of planes as I have concluded that the delay rate of planes increases and peaks when a plane is at 25 years of service, and decreases thereafter.
Factor 4: CRSElapsedTime
crs_elapsed <- main[c(1:100000),] ## Taking a proportion of data
crs_elapsed <- crs_elapsed %>%
drop_na(CRSElapsedTime)
# Scatterplot of ArrDelay against DepDelay
ggplot(crs_elapsed, aes(x=DepDelay, y=CRSElapsedTime)) +
geom_point(alpha = 0.2, size = 0.3) +
labs(title = "Scatter Plot of Scheduled Elapsed Time against Departure Delay (minutes)",
x = "Departure Delay", y = "Scheduled Elapsed Time") +
theme_classic() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(plot.title = element_text(face = "bold", hjust=0.5, vjust = 1, size = 12)) 
Conclusion: CRSDepTime, CRSArrTime,
age (of plane), CRSElapsedTime are variables
that will be included.
Plane manufacturer can’t be used as it contains too many NA’s, imputing large proportion of dataset will cause inaccuracy.
Cleaning up variables and creating new dataframe
original
original_plane <- planes %>%
select(tailnum, year)
dep = subset(dep, select = c("TailNum", "Year", "CRSDepTime", "DepDelay", "CRSElapsedTime"))
arr = subset(arr, select = c("TailNum", "Year", "CRSArrTime", "DepDelay", "CRSElapsedTime"))
original <- dep %>%
inner_join(arr)## Joining, by = c("TailNum", "Year", "DepDelay", "CRSElapsedTime")colnames(original)[1] <- "tailnum"
original <- original %>%
inner_join(original_plane)## Joining, by = "tailnum"str(original)## 'data.frame': 76736905 obs. of 7 variables:
## $ tailnum : chr "N427UA" "N449UA" "N449UA" "N433UA" ...
## $ Year : Factor w/ 3 levels "2005","2006",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ CRSDepTime : Factor w/ 24 levels "0100","0200",..: 8 8 8 8 8 8 6 6 6 6 ...
## $ DepDelay : int -4 0 0 -6 13 2 14 3 -4 -5 ...
## $ CRSElapsedTime: int 197 197 197 197 197 197 112 112 112 112 ...
## $ CRSArrTime : Factor w/ 24 levels "0100","0200",..: 11 11 16 11 11 11 7 7 7 7 ...
## $ year : num 1995 1998 1998 1996 1995 ...original$Year <- as.numeric(as.character(original$Year))
original$year <- as.numeric(original$year)
summary(original) # year(plane) has '0' and several NA's## tailnum Year CRSDepTime DepDelay
## Length:76736905 Min. :2005 0800 : 6225911 Min. :-1200.00
## Class :character 1st Qu.:2005 0700 : 5748525 1st Qu.: -4.00
## Mode :character Median :2006 1000 : 5502320 Median : 0.00
## Mean :2006 0600 : 5491855 Mean : 1.18
## 3rd Qu.:2007 0900 : 5209006 3rd Qu.: 0.00
## Max. :2007 1200 : 5096041 Max. : 2601.00
## (Other):43463247
## CRSElapsedTime CRSArrTime year
## Min. : -97.00 1000 : 6101454 Min. : 0
## 1st Qu.: 60.00 0800 : 5837636 1st Qu.:1994
## Median : 75.00 1200 : 5645913 Median :1997
## Mean : 86.89 0900 : 5303718 Mean :1994
## 3rd Qu.: 100.00 1400 : 5091362 3rd Qu.:2001
## Max. :1031.00 1600 : 5084999 Max. :2007
## NA's :1 (Other):43671823# Imputing median value
original$year[is.na(original$year)] = median(original$year, na.rm=TRUE)
original$CRSElapsedTime[is.na(original$CRSElapsedTime)] = median(original$CRSElapsedTime, na.rm=TRUE)
original <- original %>%
filter(year > 0) %>%
mutate(age = Year - year) %>%
filter(age > 0) # There will be several negative values, also a small proportion, cleaning up data
# Encoding factors
original <- original %>%
mutate_if(is.character, as.factor)
original <- original %>%
mutate_if(is.factor, as.numeric)
summary(original)## tailnum Year CRSDepTime DepDelay
## Min. : 1 Min. :2005 Min. : 1.00 Min. :-1200.000
## 1st Qu.:1618 1st Qu.:2005 1st Qu.: 9.00 1st Qu.: -4.000
## Median :2357 Median :2006 Median :12.00 Median : 0.000
## Mean :2229 Mean :2006 Mean :12.72 Mean : 1.159
## 3rd Qu.:2798 3rd Qu.:2007 3rd Qu.:17.00 3rd Qu.: 0.000
## Max. :4171 Max. :2007 Max. :24.00 Max. : 2601.000
## CRSElapsedTime CRSArrTime year age
## Min. : -97.00 Min. : 1.00 Min. :1956 Min. : 1.000
## 1st Qu.: 60.00 1st Qu.:10.00 1st Qu.:1994 1st Qu.: 5.000
## Median : 75.00 Median :14.00 Median :1997 Median : 8.000
## Mean : 86.62 Mean :14.07 Mean :1996 Mean : 9.768
## 3rd Qu.: 100.00 3rd Qu.:18.00 3rd Qu.:2001 3rd Qu.:13.000
## Max. :1031.00 Max. :24.00 Max. :2006 Max. :51.000# Checking for missing values
colSums(is.na(original)|original=='')## tailnum Year CRSDepTime DepDelay CRSElapsedTime
## 0 0 0 0 0
## CRSArrTime year age
## 0 0 0str(original) ## 'data.frame': 76010596 obs. of 8 variables:
## $ tailnum : num 1496 1606 1606 1524 1490 ...
## $ Year : num 2005 2005 2005 2005 2005 ...
## $ CRSDepTime : num 8 8 8 8 8 8 6 6 6 6 ...
## $ DepDelay : int -4 0 0 -6 13 2 14 3 -4 -5 ...
## $ CRSElapsedTime: num 197 197 197 197 197 197 112 112 112 112 ...
## $ CRSArrTime : num 11 11 16 11 11 11 7 7 7 7 ...
## $ year : num 1995 1998 1998 1996 1995 ...
## $ age : num 10 7 7 9 10 7 14 13 14 17 ...# Subsetting to columns needed
original <- original[c(3:6,8)]
str(original)## 'data.frame': 76010596 obs. of 5 variables:
## $ CRSDepTime : num 8 8 8 8 8 8 6 6 6 6 ...
## $ DepDelay : int -4 0 0 -6 13 2 14 3 -4 -5 ...
## $ CRSElapsedTime: num 197 197 197 197 197 197 112 112 112 112 ...
## $ CRSArrTime : num 11 11 16 11 11 11 7 7 7 7 ...
## $ age : num 10 7 7 9 10 7 14 13 14 17 ...Training and evaluating
task_original <- TaskRegr$new(original, id = "Delay", target = "DepDelay")
print(task_original)## <TaskRegr:Delay> (76010596 x 5)
## * Target: DepDelay
## * Properties: -
## * Features (4):
## - dbl (4): CRSArrTime, CRSDepTime, CRSElapsedTime, agetask_original$feature_names## [1] "CRSArrTime" "CRSDepTime" "CRSElapsedTime" "age"task_original$target_names## [1] "DepDelay"task_original$filter(rows = 1:100000)
task_original$select(setdiff(task_original$feature_names, "DepDelay"))
measure <- msr("regr.mse")
learner_lm <- lrn("regr.lm")
gr_lm <- po("imputemean") %>>%
po(learner_lm)
glrn_lm <- GraphLearner$new(gr_lm)
set.seed(1)
train_set <- sample(task_original$nrow, 0.7 * task_original$nrow)
test_set <- setdiff(seq_len(task_original$nrow), train_set)
glrn_lm$train(task_original, row_ids = train_set)
glrn_lm$predict(task_original, row_ids = test_set)$score()## regr.mse
## 78.64405### regr.mse = 70.12866Ridge Regression: Tuning Hyperparameters
learner_ridge <- lrn("regr.glmnet")
learner_ridge$param_set$values <- list(alpha = 0, lambda = 0.001)
gr_ridge <- po("scale") %>>%
po("imputemean") %>>%
po(learner_ridge)
glrn_ridge <- GraphLearner$new(gr_ridge)
glrn_ridge$train(task_original, row_ids = train_set)
glrn_ridge$predict(task_original, row_ids = test_set)$score()## regr.mse
## 78.64404### regr.mse = 70.12547
learner_ridge2 <- lrn("regr.glmnet")
learner_ridge2$param_set$values <- list(alpha = 0)
gr_ridge2 <- po("scale") %>>%
po("imputemean") %>>%
po(learner_ridge2)
glrn_ridge2 <- GraphLearner$new(gr_ridge2)
tune_lambda <- ParamSet$new (list(
ParamDbl$new("regr.glmnet.lambda", lower = 0.03, upper = 2)
))
tuner <- tnr("grid_search")
terminator <- trm("evals", n_evals = 20)
at_ridge <- AutoTuner$new(
learner = glrn_ridge2,
resampling = rsmp("cv", folds = 3),
measure = measure,
search_space = tune_lambda,
terminator = terminator,
tuner = tuner
)
at_ridge$train(task_original, row_ids = train_set)## INFO [04:04:55.837] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [04:04:55.938] [bbotk] Evaluating 1 configuration(s)
## INFO [04:04:56.069] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:04:56.232] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:05:12.954] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:05:26.835] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:05:40.975] [mlr3] Finished benchmark
## INFO [04:05:41.182] [bbotk] Result of batch 1:
## INFO [04:05:41.206] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:05:41.206] [bbotk] 0.03 101.3591 0 0 44.568
## INFO [04:05:41.206] [bbotk] uhash
## INFO [04:05:41.206] [bbotk] 1bb99e62-a930-437e-9cb9-1683654c1b8a
## INFO [04:05:41.214] [bbotk] Evaluating 1 configuration(s)
## INFO [04:05:41.399] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:05:41.431] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:05:54.924] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:06:08.061] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:06:19.847] [mlr3] Finished benchmark
## INFO [04:06:19.903] [bbotk] Result of batch 2:
## INFO [04:06:19.905] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:06:19.905] [bbotk] 1.124444 101.368 0 0 38.343
## INFO [04:06:19.905] [bbotk] uhash
## INFO [04:06:19.905] [bbotk] e3badafa-2a97-4b7b-b141-75534e353e67
## INFO [04:06:19.907] [bbotk] Evaluating 1 configuration(s)
## INFO [04:06:19.975] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:06:19.985] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:06:35.305] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:06:48.452] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:07:04.378] [mlr3] Finished benchmark
## INFO [04:07:04.455] [bbotk] Result of batch 3:
## INFO [04:07:04.459] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:07:04.459] [bbotk] 2 101.3841 0 0 44.323
## INFO [04:07:04.459] [bbotk] uhash
## INFO [04:07:04.459] [bbotk] fb80b5f5-470b-4c17-9cb9-f6936a4f77e8
## INFO [04:07:04.461] [bbotk] Evaluating 1 configuration(s)
## INFO [04:07:04.562] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:07:04.578] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:07:17.643] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:07:31.230] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:07:44.274] [mlr3] Finished benchmark
## INFO [04:07:44.433] [bbotk] Result of batch 4:
## INFO [04:07:44.438] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:07:44.438] [bbotk] 0.9055556 101.3651 0 0 39.624
## INFO [04:07:44.438] [bbotk] uhash
## INFO [04:07:44.438] [bbotk] 0fa39165-eab7-4739-95da-09a421dde0dc
## INFO [04:07:44.443] [bbotk] Evaluating 1 configuration(s)
## INFO [04:07:44.517] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:07:44.528] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:07:56.502] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:08:15.630] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:08:31.694] [mlr3] Finished benchmark
## INFO [04:08:31.868] [bbotk] Result of batch 5:
## INFO [04:08:31.873] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:08:31.873] [bbotk] 1.781111 101.3795 0 0 47.062
## INFO [04:08:31.873] [bbotk] uhash
## INFO [04:08:31.873] [bbotk] e9fbbeae-17c3-4a36-96b5-cc353e752237
## INFO [04:08:31.880] [bbotk] Evaluating 1 configuration(s)
## INFO [04:08:32.021] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:08:32.040] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:08:47.770] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:09:02.833] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:09:16.761] [mlr3] Finished benchmark
## INFO [04:09:16.884] [bbotk] Result of batch 6:
## INFO [04:09:16.888] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:09:16.888] [bbotk] 0.4677778 101.3607 0 0 44.606
## INFO [04:09:16.888] [bbotk] uhash
## INFO [04:09:16.888] [bbotk] 7dd216bc-59e2-465f-9ac4-b26bd817dba1
## INFO [04:09:16.892] [bbotk] Evaluating 1 configuration(s)
## INFO [04:09:17.002] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:09:17.021] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:09:32.574] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:09:45.321] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:10:05.538] [mlr3] Finished benchmark
## INFO [04:10:05.708] [bbotk] Result of batch 7:
## INFO [04:10:05.713] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:10:05.713] [bbotk] 1.343333 101.3715 0 0 48.366
## INFO [04:10:05.713] [bbotk] uhash
## INFO [04:10:05.713] [bbotk] 08a19f0b-4ebc-44cc-8448-9c30988dcabb
## INFO [04:10:05.717] [bbotk] Evaluating 1 configuration(s)
## INFO [04:10:05.822] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:10:05.840] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:10:21.067] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:10:35.523] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:10:49.160] [mlr3] Finished benchmark
## INFO [04:10:49.243] [bbotk] Result of batch 8:
## INFO [04:10:49.246] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:10:49.246] [bbotk] 0.2488889 101.3595 0 0 43.264
## INFO [04:10:49.246] [bbotk] uhash
## INFO [04:10:49.246] [bbotk] 44902c76-c509-472c-be50-cc5581639d3b
## INFO [04:10:49.248] [bbotk] Evaluating 1 configuration(s)
## INFO [04:10:49.312] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:10:49.321] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:11:03.151] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:11:18.961] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:11:32.642] [mlr3] Finished benchmark
## INFO [04:11:32.771] [bbotk] Result of batch 9:
## INFO [04:11:32.775] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:11:32.775] [bbotk] 1.562222 101.3753 0 0 43.207
## INFO [04:11:32.775] [bbotk] uhash
## INFO [04:11:32.775] [bbotk] 47d1ecd9-10d2-47e1-89e2-85ed4cfcb4f2
## INFO [04:11:32.778] [bbotk] Evaluating 1 configuration(s)
## INFO [04:11:32.878] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:11:32.895] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:11:49.071] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:12:03.173] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:12:28.475] [mlr3] Finished benchmark
## INFO [04:12:28.537] [bbotk] Result of batch 10:
## INFO [04:12:28.539] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:12:28.539] [bbotk] 0.6866667 101.3626 0 0 55.502
## INFO [04:12:28.539] [bbotk] uhash
## INFO [04:12:28.539] [bbotk] 45892faf-acb2-4e5e-b831-f0c892d5d99d
## INFO [04:12:28.547] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [04:12:28.548] [bbotk] Result:
## INFO [04:12:28.550] [bbotk] regr.glmnet.lambda learner_param_vals x_domain regr.mse
## INFO [04:12:28.550] [bbotk] 0.03 <list[3]> <list[1]> 101.3591at_ridge$predict(task_original, row_ids = test_set)$score()## regr.mse
## 78.64359### regr.mse = 70.13002Random Forests
learner_rf <- lrn('regr.ranger')
learner_rf$param_set$values <- list(min.node.size = 4)
gr_rf <- po('scale') %>>%
po('imputemean') %>>%
po(learner_rf)
glrn_rf <- GraphLearner$new(gr_rf)
tune_ntrees <- ParamSet$new (list(
ParamInt$new('regr.ranger.num.trees', lower = 50, upper = 600)))
at_rf <- AutoTuner$new(
learner = glrn_rf,
resampling = rsmp('cv', folds = 3),
measure = measure,
search_space = tune_ntrees,
terminator = terminator,
tuner = tuner)
at_rf$train(task_original, row_ids = train_set)## INFO [04:12:48.867] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [04:12:48.891] [bbotk] Evaluating 1 configuration(s)
## INFO [04:12:49.015] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:12:49.045] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:13:10.142] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:13:30.389] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:13:52.395] [mlr3] Finished benchmark
## INFO [04:13:52.496] [bbotk] Result of batch 1:
## INFO [04:13:52.503] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:13:52.503] [bbotk] 172 106.9784 0 0 63.184
## INFO [04:13:52.503] [bbotk] uhash
## INFO [04:13:52.503] [bbotk] 8c6048fb-6bd9-46e8-89ce-022feed544ff
## INFO [04:13:52.506] [bbotk] Evaluating 1 configuration(s)
## INFO [04:13:52.605] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:13:52.625] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:14:22.159] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:14:51.381] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:15:23.481] [mlr3] Finished benchmark
## INFO [04:15:23.893] [bbotk] Result of batch 2:
## INFO [04:15:23.916] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:15:23.916] [bbotk] 600 106.7579 0 0 90.4
## INFO [04:15:23.916] [bbotk] uhash
## INFO [04:15:23.916] [bbotk] bfe83a56-1831-488a-9395-6565e76c5478
## INFO [04:15:23.930] [bbotk] Evaluating 1 configuration(s)
## INFO [04:15:24.228] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:15:24.299] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:15:51.014] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:16:17.163] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:16:41.515] [mlr3] Finished benchmark
## INFO [04:16:41.874] [bbotk] Result of batch 3:
## INFO [04:16:41.890] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:16:41.890] [bbotk] 356 107.1674 0 0 76.932
## INFO [04:16:41.890] [bbotk] uhash
## INFO [04:16:41.890] [bbotk] 12586678-1444-4d0a-88b8-74ed5b084863
## INFO [04:16:41.897] [bbotk] Evaluating 1 configuration(s)
## INFO [04:16:42.197] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:16:42.237] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:17:09.079] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:17:34.603] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:18:00.931] [mlr3] Finished benchmark
## INFO [04:18:01.153] [bbotk] Result of batch 4:
## INFO [04:18:01.163] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:18:01.163] [bbotk] 478 107.0987 0 0 78.531
## INFO [04:18:01.163] [bbotk] uhash
## INFO [04:18:01.163] [bbotk] 7fa40647-5990-434d-bac4-d568070b3847
## INFO [04:18:01.177] [bbotk] Evaluating 1 configuration(s)
## INFO [04:18:01.372] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:18:01.407] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:18:22.273] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:18:42.466] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:19:03.414] [mlr3] Finished benchmark
## INFO [04:19:03.620] [bbotk] Result of batch 5:
## INFO [04:19:03.628] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:19:03.628] [bbotk] 233 106.9801 0 0 61.872
## INFO [04:19:03.628] [bbotk] uhash
## INFO [04:19:03.628] [bbotk] 910106cf-c31d-4b96-93e8-4aa55dced883
## INFO [04:19:03.632] [bbotk] Evaluating 1 configuration(s)
## INFO [04:19:03.801] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:19:03.821] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:19:32.984] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:19:58.931] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:20:27.035] [mlr3] Finished benchmark
## INFO [04:20:27.270] [bbotk] Result of batch 6:
## INFO [04:20:27.275] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:20:27.275] [bbotk] 539 107.0493 0 0 83.052
## INFO [04:20:27.275] [bbotk] uhash
## INFO [04:20:27.275] [bbotk] cdd2dab4-7764-4d52-9274-c4025d30d9c2
## INFO [04:20:27.279] [bbotk] Evaluating 1 configuration(s)
## INFO [04:20:27.413] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:20:27.431] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:20:48.651] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:21:10.526] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:21:32.226] [mlr3] Finished benchmark
## INFO [04:21:32.413] [bbotk] Result of batch 7:
## INFO [04:21:32.418] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:21:32.418] [bbotk] 294 107.1791 0 0 64.646
## INFO [04:21:32.418] [bbotk] uhash
## INFO [04:21:32.418] [bbotk] 4221803d-d8ca-42d1-8490-c831d61202e0
## INFO [04:21:32.423] [bbotk] Evaluating 1 configuration(s)
## INFO [04:21:32.552] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:21:32.574] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:21:57.104] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:22:21.523] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:22:45.686] [mlr3] Finished benchmark
## INFO [04:22:45.895] [bbotk] Result of batch 8:
## INFO [04:22:45.903] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:22:45.903] [bbotk] 417 107.1271 0 0 72.987
## INFO [04:22:45.903] [bbotk] uhash
## INFO [04:22:45.903] [bbotk] 57b7eb28-5980-4847-a2a7-44af08aca112
## INFO [04:22:45.907] [bbotk] Evaluating 1 configuration(s)
## INFO [04:22:46.066] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:22:46.094] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:23:03.924] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:23:20.903] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:23:37.687] [mlr3] Finished benchmark
## INFO [04:23:37.863] [bbotk] Result of batch 9:
## INFO [04:23:37.869] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:23:37.869] [bbotk] 111 107.1279 0 0 51.471
## INFO [04:23:37.869] [bbotk] uhash
## INFO [04:23:37.869] [bbotk] b471d40b-1673-41b6-ab52-fbd972c8f5a6
## INFO [04:23:37.874] [bbotk] Evaluating 1 configuration(s)
## INFO [04:23:38.023] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:23:38.040] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:23:54.461] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:24:10.505] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:24:27.269] [mlr3] Finished benchmark
## INFO [04:24:27.452] [bbotk] Result of batch 10:
## INFO [04:24:27.458] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:24:27.458] [bbotk] 50 107.7743 0 0 49.118
## INFO [04:24:27.458] [bbotk] uhash
## INFO [04:24:27.458] [bbotk] b9b7f423-d2ae-42b8-a9af-b66c2adfb2de
## INFO [04:24:27.543] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [04:24:27.548] [bbotk] Result:
## INFO [04:24:27.553] [bbotk] regr.ranger.num.trees learner_param_vals x_domain regr.mse
## INFO [04:24:27.553] [bbotk] 600 <list[3]> <list[1]> 106.7579at_rf$predict(task_original, row_ids = test_set)$score()## regr.mse
## 83.88102### regr.mse = 35.71414Benchmarking
set.seed(123)
# list of learners
lrn_list <- list(
glrn_lm,
glrn_ridge,
at_ridge,
at_rf
)
# Set the benchmark design and run the comparisons
bm_design <- benchmark_grid(task = task_original, resamplings = rsmp('cv', folds = 3),
learners = lrn_list)
bmr <- benchmark(bm_design, store_models = TRUE)## INFO [04:25:06.391] [mlr3] Running benchmark with 12 resampling iterations
## INFO [04:25:06.411] [mlr3] Applying learner 'imputemean.regr.lm' on task 'Delay' (iter 1/3)
## INFO [04:25:12.588] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:25:27.166] [mlr3] Applying learner 'scale.imputemean.regr.ranger.tuned' on task 'Delay' (iter 2/3)
## INFO [04:25:27.895] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [04:25:27.916] [bbotk] Evaluating 1 configuration(s)
## INFO [04:25:28.129] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:25:28.165] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:25:45.221] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:26:00.662] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:26:16.352] [mlr3] Finished benchmark
## INFO [04:26:16.553] [bbotk] Result of batch 1:
## INFO [04:26:16.560] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:26:16.560] [bbotk] 50 101.7081 0 0 48.02
## INFO [04:26:16.560] [bbotk] uhash
## INFO [04:26:16.560] [bbotk] db5e87ed-c85d-4606-b16e-6fe6d65878df
## INFO [04:26:16.566] [bbotk] Evaluating 1 configuration(s)
## INFO [04:26:16.786] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:26:16.812] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:26:37.511] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:26:56.552] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:27:17.484] [mlr3] Finished benchmark
## INFO [04:27:17.660] [bbotk] Result of batch 2:
## INFO [04:27:17.673] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:27:17.673] [bbotk] 294 101.2209 0 0 60.577
## INFO [04:27:17.673] [bbotk] uhash
## INFO [04:27:17.673] [bbotk] c96c2a83-726a-465d-a076-49079309b3ad
## INFO [04:27:17.676] [bbotk] Evaluating 1 configuration(s)
## INFO [04:27:17.834] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:27:17.880] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:27:34.514] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:27:51.515] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:28:08.070] [mlr3] Finished benchmark
## INFO [04:28:08.133] [bbotk] Result of batch 3:
## INFO [04:28:08.136] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:28:08.136] [bbotk] 111 101.6412 0 0 50.096
## INFO [04:28:08.136] [bbotk] uhash
## INFO [04:28:08.136] [bbotk] e18ecc07-6b72-4aa6-831b-7d89bcf1f716
## INFO [04:28:08.138] [bbotk] Evaluating 1 configuration(s)
## INFO [04:28:08.189] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:28:08.205] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:28:32.206] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:28:58.013] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:29:21.826] [mlr3] Finished benchmark
## INFO [04:29:22.010] [bbotk] Result of batch 4:
## INFO [04:29:22.016] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:29:22.016] [bbotk] 539 100.9735 0 0 73.525
## INFO [04:29:22.016] [bbotk] uhash
## INFO [04:29:22.016] [bbotk] 05dbbba1-c650-412d-983e-7a4ea212a375
## INFO [04:29:22.022] [bbotk] Evaluating 1 configuration(s)
## INFO [04:29:22.286] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:29:22.344] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:29:44.256] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:30:05.663] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:30:26.298] [mlr3] Finished benchmark
## INFO [04:30:26.362] [bbotk] Result of batch 5:
## INFO [04:30:26.364] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:30:26.364] [bbotk] 356 101.1191 0 0 63.891
## INFO [04:30:26.364] [bbotk] uhash
## INFO [04:30:26.364] [bbotk] 18f4a04a-714d-4a23-9389-8b1955f225b3
## INFO [04:30:26.366] [bbotk] Evaluating 1 configuration(s)
## INFO [04:30:26.409] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:30:26.418] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:30:49.024] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:31:12.113] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:31:42.188] [mlr3] Finished benchmark
## INFO [04:31:42.276] [bbotk] Result of batch 6:
## INFO [04:31:42.279] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:31:42.279] [bbotk] 478 101.0684 0 0 75.697
## INFO [04:31:42.279] [bbotk] uhash
## INFO [04:31:42.279] [bbotk] b5c8e22c-f7b5-454a-ae9f-5d16fb038e80
## INFO [04:31:42.281] [bbotk] Evaluating 1 configuration(s)
## INFO [04:31:42.335] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:31:42.347] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:32:13.135] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:32:41.323] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:33:09.826] [mlr3] Finished benchmark
## INFO [04:33:10.269] [bbotk] Result of batch 7:
## INFO [04:33:10.307] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:33:10.307] [bbotk] 600 100.9992 0 0 87.171
## INFO [04:33:10.307] [bbotk] uhash
## INFO [04:33:10.307] [bbotk] 2737bee5-bf83-4074-9f46-0776f3a817f1
## INFO [04:33:10.315] [bbotk] Evaluating 1 configuration(s)
## INFO [04:33:10.591] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:33:10.647] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:33:31.269] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:33:52.609] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:34:12.103] [mlr3] Finished benchmark
## INFO [04:34:12.515] [bbotk] Result of batch 8:
## INFO [04:34:12.533] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:34:12.533] [bbotk] 233 101.1252 0 0 61.248
## INFO [04:34:12.533] [bbotk] uhash
## INFO [04:34:12.533] [bbotk] fe38a62b-7532-448e-b40f-ffc50393a03e
## INFO [04:34:12.541] [bbotk] Evaluating 1 configuration(s)
## INFO [04:34:12.815] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:34:12.849] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:34:41.311] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:35:07.207] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:35:31.000] [mlr3] Finished benchmark
## INFO [04:35:31.302] [bbotk] Result of batch 9:
## INFO [04:35:31.314] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:35:31.314] [bbotk] 172 101.2901 0 0 77.886
## INFO [04:35:31.314] [bbotk] uhash
## INFO [04:35:31.314] [bbotk] b69669d7-474d-4a46-a2cb-b8729293b445
## INFO [04:35:31.321] [bbotk] Evaluating 1 configuration(s)
## INFO [04:35:31.554] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:35:31.583] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:36:05.160] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:36:36.812] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:37:09.519] [mlr3] Finished benchmark
## INFO [04:37:09.810] [bbotk] Result of batch 10:
## INFO [04:37:09.820] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:37:09.820] [bbotk] 417 101.2634 0 0 97.611
## INFO [04:37:09.820] [bbotk] uhash
## INFO [04:37:09.820] [bbotk] ae570db2-a02a-4913-8419-9f8242ad0c3b
## INFO [04:37:09.875] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [04:37:09.877] [bbotk] Result:
## INFO [04:37:09.883] [bbotk] regr.ranger.num.trees learner_param_vals x_domain regr.mse
## INFO [04:37:09.883] [bbotk] 539 <list[3]> <list[1]> 100.9735
## INFO [04:38:01.966] [mlr3] Applying learner 'imputemean.regr.lm' on task 'Delay' (iter 3/3)
## INFO [04:38:11.582] [mlr3] Applying learner 'imputemean.regr.lm' on task 'Delay' (iter 2/3)
## INFO [04:38:19.873] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:38:40.024] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:38:57.149] [mlr3] Applying learner 'scale.imputemean.regr.ranger.tuned' on task 'Delay' (iter 3/3)
## INFO [04:38:58.322] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [04:38:58.345] [bbotk] Evaluating 1 configuration(s)
## INFO [04:38:58.510] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:38:58.552] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:39:38.406] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:40:13.366] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:40:43.885] [mlr3] Finished benchmark
## INFO [04:40:44.240] [bbotk] Result of batch 1:
## INFO [04:40:44.261] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:40:44.261] [bbotk] 600 93.1543 0 0 104.869
## INFO [04:40:44.261] [bbotk] uhash
## INFO [04:40:44.261] [bbotk] 4e6f5d13-7e37-4e68-96cf-da7795394ccd
## INFO [04:40:44.268] [bbotk] Evaluating 1 configuration(s)
## INFO [04:40:44.415] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:40:44.442] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:41:08.574] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:41:43.022] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:42:19.044] [mlr3] Finished benchmark
## INFO [04:42:19.508] [bbotk] Result of batch 2:
## INFO [04:42:19.524] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:42:19.524] [bbotk] 233 93.49332 0 0 93.993
## INFO [04:42:19.524] [bbotk] uhash
## INFO [04:42:19.524] [bbotk] a2bb6f7f-2dfc-4c7c-b5c7-8ee09b415246
## INFO [04:42:19.534] [bbotk] Evaluating 1 configuration(s)
## INFO [04:42:19.815] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:42:19.888] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:42:52.893] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:43:33.623] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:44:19.392] [mlr3] Finished benchmark
## INFO [04:44:20.826] [bbotk] Result of batch 3:
## INFO [04:44:20.853] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:44:20.853] [bbotk] 356 93.22086 0 0 118.789
## INFO [04:44:20.853] [bbotk] uhash
## INFO [04:44:20.853] [bbotk] 359e4b6f-b74a-40e8-b2b8-c77f4b29f3d3
## INFO [04:44:20.866] [bbotk] Evaluating 1 configuration(s)
## INFO [04:44:21.198] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:44:21.264] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:45:14.562] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:47:23.522] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:48:01.467] [mlr3] Finished benchmark
## INFO [04:48:02.519] [bbotk] Result of batch 4:
## INFO [04:48:02.543] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:48:02.543] [bbotk] 294 93.17571 0 0 219.125
## INFO [04:48:02.543] [bbotk] uhash
## INFO [04:48:02.543] [bbotk] 6839a2d9-6d1b-4b73-84f5-a48df088f98e
## INFO [04:48:02.552] [bbotk] Evaluating 1 configuration(s)
## INFO [04:48:02.777] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:48:02.822] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:48:23.131] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:48:42.226] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:48:57.805] [mlr3] Finished benchmark
## INFO [04:48:58.047] [bbotk] Result of batch 5:
## INFO [04:48:58.052] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:48:58.052] [bbotk] 50 93.87929 0 0 54.801
## INFO [04:48:58.052] [bbotk] uhash
## INFO [04:48:58.052] [bbotk] 08c18333-fa85-4ebd-ba49-d4c01291bad2
## INFO [04:48:58.055] [bbotk] Evaluating 1 configuration(s)
## INFO [04:48:58.123] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:48:58.137] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:49:24.509] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:49:55.676] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:50:38.863] [mlr3] Finished benchmark
## INFO [04:50:39.202] [bbotk] Result of batch 6:
## INFO [04:50:39.224] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:50:39.224] [bbotk] 417 93.39374 0 0 100.426
## INFO [04:50:39.224] [bbotk] uhash
## INFO [04:50:39.224] [bbotk] b8e3cff2-12b4-4c47-a429-729c391c17ed
## INFO [04:50:39.234] [bbotk] Evaluating 1 configuration(s)
## INFO [04:50:39.566] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:50:39.604] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:51:03.838] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:51:27.571] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:51:48.030] [mlr3] Finished benchmark
## INFO [04:51:48.276] [bbotk] Result of batch 7:
## INFO [04:51:48.293] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:51:48.293] [bbotk] 172 93.09661 0 0 68.128
## INFO [04:51:48.293] [bbotk] uhash
## INFO [04:51:48.293] [bbotk] 4ea6fe48-87d4-4523-9d6f-4b15c9bed0f6
## INFO [04:51:48.301] [bbotk] Evaluating 1 configuration(s)
## INFO [04:51:48.512] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:51:48.545] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:52:15.536] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:52:32.574] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:52:54.544] [mlr3] Finished benchmark
## INFO [04:52:54.893] [bbotk] Result of batch 8:
## INFO [04:52:54.914] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:52:54.914] [bbotk] 111 93.51544 0 0 65.826
## INFO [04:52:54.914] [bbotk] uhash
## INFO [04:52:54.914] [bbotk] 8a4d09be-00b2-4aa3-9918-1600289f26a5
## INFO [04:52:54.921] [bbotk] Evaluating 1 configuration(s)
## INFO [04:52:55.154] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:52:55.214] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:53:24.946] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:54:01.465] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:54:35.387] [mlr3] Finished benchmark
## INFO [04:54:35.669] [bbotk] Result of batch 9:
## INFO [04:54:35.679] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:54:35.679] [bbotk] 478 93.33387 0 0 99.846
## INFO [04:54:35.679] [bbotk] uhash
## INFO [04:54:35.679] [bbotk] 14c3442b-9319-4cc9-b19b-62f4dbd527ab
## INFO [04:54:35.683] [bbotk] Evaluating 1 configuration(s)
## INFO [04:54:35.785] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:54:35.799] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [04:55:10.709] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [04:55:43.360] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [04:56:18.532] [mlr3] Finished benchmark
## INFO [04:56:19.111] [bbotk] Result of batch 10:
## INFO [04:56:19.125] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [04:56:19.125] [bbotk] 539 93.05796 0 0 102.395
## INFO [04:56:19.125] [bbotk] uhash
## INFO [04:56:19.125] [bbotk] 2cf23859-575c-4bc2-a87f-0328524f3870
## INFO [04:56:19.209] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [04:56:19.211] [bbotk] Result:
## INFO [04:56:19.218] [bbotk] regr.ranger.num.trees learner_param_vals x_domain regr.mse
## INFO [04:56:19.218] [bbotk] 539 <list[3]> <list[1]> 93.05796
## INFO [04:57:02.019] [mlr3] Applying learner 'scale.imputemean.regr.glmnet.tuned' on task 'Delay' (iter 2/3)
## INFO [04:57:02.742] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [04:57:02.762] [bbotk] Evaluating 1 configuration(s)
## INFO [04:57:02.872] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:57:02.898] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:57:19.966] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:57:37.031] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:58:00.419] [mlr3] Finished benchmark
## INFO [04:58:00.719] [bbotk] Result of batch 1:
## INFO [04:58:00.732] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:58:00.732] [bbotk] 1.562222 98.23961 0 0 57.197
## INFO [04:58:00.732] [bbotk] uhash
## INFO [04:58:00.732] [bbotk] 18f408e8-7349-4d76-ae8f-0f0ca7838c17
## INFO [04:58:00.740] [bbotk] Evaluating 1 configuration(s)
## INFO [04:58:01.103] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:58:01.227] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:58:16.544] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:58:32.435] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:58:46.364] [mlr3] Finished benchmark
## INFO [04:58:46.595] [bbotk] Result of batch 2:
## INFO [04:58:46.613] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:58:46.613] [bbotk] 0.2488889 98.2254 0 0 44.969
## INFO [04:58:46.613] [bbotk] uhash
## INFO [04:58:46.613] [bbotk] a335b023-87bd-497c-afb4-ae606d0f639a
## INFO [04:58:46.624] [bbotk] Evaluating 1 configuration(s)
## INFO [04:58:49.867] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:58:49.970] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [04:59:20.121] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [04:59:40.484] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [04:59:58.924] [mlr3] Finished benchmark
## INFO [04:59:59.315] [bbotk] Result of batch 3:
## INFO [04:59:59.331] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [04:59:59.331] [bbotk] 0.4677778 98.22634 0 0 68.614
## INFO [04:59:59.331] [bbotk] uhash
## INFO [04:59:59.331] [bbotk] 66bf0b6f-0601-4759-9148-2a5e90c53372
## INFO [04:59:59.340] [bbotk] Evaluating 1 configuration(s)
## INFO [04:59:59.610] [mlr3] Running benchmark with 3 resampling iterations
## INFO [04:59:59.658] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:00:17.608] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:00:31.310] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:00:51.720] [mlr3] Finished benchmark
## INFO [05:00:52.205] [bbotk] Result of batch 4:
## INFO [05:00:52.222] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:00:52.222] [bbotk] 1.781111 98.24353 0 0 51.874
## INFO [05:00:52.222] [bbotk] uhash
## INFO [05:00:52.222] [bbotk] c7580234-4fbd-4d5d-9ea7-e758b0dc8097
## INFO [05:00:52.231] [bbotk] Evaluating 1 configuration(s)
## INFO [05:00:52.506] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:00:52.545] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:01:09.423] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:01:34.678] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:01:54.785] [mlr3] Finished benchmark
## INFO [05:01:55.258] [bbotk] Result of batch 5:
## INFO [05:01:55.272] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:01:55.272] [bbotk] 1.343333 98.23604 0 0 61.912
## INFO [05:01:55.272] [bbotk] uhash
## INFO [05:01:55.272] [bbotk] a6759be1-e5b2-42b9-a604-4a4b22310c75
## INFO [05:01:55.292] [bbotk] Evaluating 1 configuration(s)
## INFO [05:01:55.641] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:01:55.694] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:03:45.585] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:04:16.709] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:04:43.324] [mlr3] Finished benchmark
## INFO [05:04:44.135] [bbotk] Result of batch 6:
## INFO [05:04:44.200] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:04:44.200] [bbotk] 2 98.24775 0 0 166.827
## INFO [05:04:44.200] [bbotk] uhash
## INFO [05:04:44.200] [bbotk] e5f1f380-8468-4812-ae6f-d4c833bb2bfa
## INFO [05:04:44.260] [bbotk] Evaluating 1 configuration(s)
## INFO [05:04:44.835] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:04:44.961] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:07:27.498] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:07:47.881] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:08:07.277] [mlr3] Finished benchmark
## INFO [05:08:07.657] [bbotk] Result of batch 7:
## INFO [05:08:07.673] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:08:07.673] [bbotk] 0.03 98.22522 0 0 201.62
## INFO [05:08:07.673] [bbotk] uhash
## INFO [05:08:07.673] [bbotk] f9756dd0-28da-4b29-9a5d-fe63298e1b67
## INFO [05:08:07.683] [bbotk] Evaluating 1 configuration(s)
## INFO [05:08:07.950] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:08:08.010] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:08:24.808] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:08:42.025] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:09:00.491] [mlr3] Finished benchmark
## INFO [05:09:00.879] [bbotk] Result of batch 8:
## INFO [05:09:00.894] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:09:00.894] [bbotk] 0.6866667 98.22795 0 0 52.195
## INFO [05:09:00.894] [bbotk] uhash
## INFO [05:09:00.894] [bbotk] 5713da6a-207f-48c0-ac21-3a4ef9fcfb08
## INFO [05:09:00.899] [bbotk] Evaluating 1 configuration(s)
## INFO [05:09:01.164] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:09:01.211] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:09:19.069] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:09:34.881] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:09:48.686] [mlr3] Finished benchmark
## INFO [05:09:48.863] [bbotk] Result of batch 9:
## INFO [05:09:48.869] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:09:48.869] [bbotk] 0.9055556 98.23015 0 0 47.219
## INFO [05:09:48.869] [bbotk] uhash
## INFO [05:09:48.869] [bbotk] 532a7faf-85e4-4052-be21-f3b07c9fa3df
## INFO [05:09:48.873] [bbotk] Evaluating 1 configuration(s)
## INFO [05:09:48.998] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:09:49.022] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:10:03.697] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:10:20.223] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:10:34.032] [mlr3] Finished benchmark
## INFO [05:10:34.476] [bbotk] Result of batch 10:
## INFO [05:10:34.486] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:10:34.486] [bbotk] 1.124444 98.23286 0 0 44.832
## INFO [05:10:34.486] [bbotk] uhash
## INFO [05:10:34.486] [bbotk] a90984fa-76d1-4b58-ad66-a6f22cf67bc6
## INFO [05:10:34.584] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [05:10:34.586] [bbotk] Result:
## INFO [05:10:34.597] [bbotk] regr.glmnet.lambda learner_param_vals x_domain regr.mse
## INFO [05:10:34.597] [bbotk] 0.03 <list[3]> <list[1]> 98.22522
## INFO [05:10:48.780] [mlr3] Applying learner 'scale.imputemean.regr.glmnet.tuned' on task 'Delay' (iter 3/3)
## INFO [05:10:49.318] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [05:10:49.331] [bbotk] Evaluating 1 configuration(s)
## INFO [05:10:49.451] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:10:49.470] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:11:04.142] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:11:22.060] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:11:36.086] [mlr3] Finished benchmark
## INFO [05:11:36.337] [bbotk] Result of batch 1:
## INFO [05:11:36.348] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:11:36.348] [bbotk] 0.4677778 89.65833 0 0 46.346
## INFO [05:11:36.348] [bbotk] uhash
## INFO [05:11:36.348] [bbotk] 8e15be5a-17b3-4821-af9a-88ddfe4aa967
## INFO [05:11:36.352] [bbotk] Evaluating 1 configuration(s)
## INFO [05:11:36.484] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:11:36.509] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:11:51.327] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:12:08.430] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:12:22.224] [mlr3] Finished benchmark
## INFO [05:12:22.429] [bbotk] Result of batch 2:
## INFO [05:12:22.438] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:12:22.438] [bbotk] 1.562222 89.67329 0 0 45.494
## INFO [05:12:22.438] [bbotk] uhash
## INFO [05:12:22.438] [bbotk] 43b8b5b5-d3e2-4439-b98e-ee15be115694
## INFO [05:12:22.443] [bbotk] Evaluating 1 configuration(s)
## INFO [05:12:22.560] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:12:22.571] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:12:38.284] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:12:54.545] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:13:09.406] [mlr3] Finished benchmark
## INFO [05:13:09.630] [bbotk] Result of batch 3:
## INFO [05:13:09.639] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:13:09.639] [bbotk] 0.6866667 89.66019 0 0 46.695
## INFO [05:13:09.639] [bbotk] uhash
## INFO [05:13:09.639] [bbotk] 14e4a2ed-3685-42a3-8e9e-bedabe9ef8b0
## INFO [05:13:09.645] [bbotk] Evaluating 1 configuration(s)
## INFO [05:13:09.765] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:13:09.782] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:13:28.831] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:13:45.223] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:13:58.363] [mlr3] Finished benchmark
## INFO [05:13:58.521] [bbotk] Result of batch 4:
## INFO [05:13:58.528] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:13:58.528] [bbotk] 1.343333 89.66931 0 0 48.406
## INFO [05:13:58.528] [bbotk] uhash
## INFO [05:13:58.528] [bbotk] 74a9d11b-cee1-4361-b93f-564b784edb44
## INFO [05:13:58.534] [bbotk] Evaluating 1 configuration(s)
## INFO [05:13:58.632] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:13:58.644] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:14:12.913] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:14:29.827] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:14:43.602] [mlr3] Finished benchmark
## INFO [05:14:43.778] [bbotk] Result of batch 5:
## INFO [05:14:43.783] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:14:43.783] [bbotk] 0.9055556 89.66269 0 0 44.811
## INFO [05:14:43.783] [bbotk] uhash
## INFO [05:14:43.783] [bbotk] f8ac3ce4-d689-4fbb-bc2a-d306fb2886de
## INFO [05:14:43.786] [bbotk] Evaluating 1 configuration(s)
## INFO [05:14:43.921] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:14:43.937] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:15:02.657] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:15:16.527] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:15:31.662] [mlr3] Finished benchmark
## INFO [05:15:31.807] [bbotk] Result of batch 6:
## INFO [05:15:31.813] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:15:31.813] [bbotk] 1.124444 89.66575 0 0 47.525
## INFO [05:15:31.813] [bbotk] uhash
## INFO [05:15:31.813] [bbotk] ffb56033-50ea-4330-8106-bc4c8e42b71c
## INFO [05:15:31.817] [bbotk] Evaluating 1 configuration(s)
## INFO [05:15:31.900] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:15:31.919] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:15:48.216] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:16:02.117] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:16:19.574] [mlr3] Finished benchmark
## INFO [05:16:19.799] [bbotk] Result of batch 7:
## INFO [05:16:19.808] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:16:19.808] [bbotk] 0.03 89.65695 0 0 47.483
## INFO [05:16:19.808] [bbotk] uhash
## INFO [05:16:19.808] [bbotk] ff4b7aa4-e975-4be0-80bf-f86e979a8a60
## INFO [05:16:19.818] [bbotk] Evaluating 1 configuration(s)
## INFO [05:16:20.114] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:16:20.167] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:16:34.736] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:16:49.734] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:17:04.785] [mlr3] Finished benchmark
## INFO [05:17:05.078] [bbotk] Result of batch 8:
## INFO [05:17:05.089] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:17:05.089] [bbotk] 0.2488889 89.65722 0 0 44.415
## INFO [05:17:05.089] [bbotk] uhash
## INFO [05:17:05.089] [bbotk] 48629a4f-4884-46f5-a623-499004f8c2d8
## INFO [05:17:05.095] [bbotk] Evaluating 1 configuration(s)
## INFO [05:17:05.244] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:17:05.256] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:17:21.816] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:17:36.777] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:17:52.340] [mlr3] Finished benchmark
## INFO [05:17:52.585] [bbotk] Result of batch 9:
## INFO [05:17:52.594] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:17:52.594] [bbotk] 2 89.68232 0 0 46.855
## INFO [05:17:52.594] [bbotk] uhash
## INFO [05:17:52.594] [bbotk] 61111ec1-bdb5-4793-9456-7c2abdb145cc
## INFO [05:17:52.601] [bbotk] Evaluating 1 configuration(s)
## INFO [05:17:52.938] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:17:53.032] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:18:06.766] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:18:20.638] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:18:42.146] [mlr3] Finished benchmark
## INFO [05:18:42.432] [bbotk] Result of batch 10:
## INFO [05:18:42.443] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:18:42.443] [bbotk] 1.781111 89.67764 0 0 48.953
## INFO [05:18:42.443] [bbotk] uhash
## INFO [05:18:42.443] [bbotk] 0519da49-8068-4310-9343-9205dfc597c3
## INFO [05:18:42.506] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [05:18:42.509] [bbotk] Result:
## INFO [05:18:42.515] [bbotk] regr.glmnet.lambda learner_param_vals x_domain regr.mse
## INFO [05:18:42.515] [bbotk] 0.03 <list[3]> <list[1]> 89.65695
## INFO [05:19:04.191] [mlr3] Applying learner 'scale.imputemean.regr.glmnet.tuned' on task 'Delay' (iter 1/3)
## INFO [05:19:05.212] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [05:19:05.219] [bbotk] Evaluating 1 configuration(s)
## INFO [05:19:05.294] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:19:05.320] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:19:31.978] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:20:27.209] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:20:54.844] [mlr3] Finished benchmark
## INFO [05:20:55.354] [bbotk] Result of batch 1:
## INFO [05:20:55.378] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:20:55.378] [bbotk] 0.03 95.76616 0 0 109.196
## INFO [05:20:55.378] [bbotk] uhash
## INFO [05:20:55.378] [bbotk] 1ca3b836-a1f1-4cce-b9fe-cc51dcfbdf13
## INFO [05:20:55.385] [bbotk] Evaluating 1 configuration(s)
## INFO [05:20:55.628] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:20:55.647] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:21:10.493] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:21:27.367] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:21:47.672] [mlr3] Finished benchmark
## INFO [05:21:48.560] [bbotk] Result of batch 2:
## INFO [05:21:48.590] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:21:48.590] [bbotk] 2 95.78784 0 0 51.551
## INFO [05:21:48.590] [bbotk] uhash
## INFO [05:21:48.590] [bbotk] 8b40c5e0-8161-4a63-a9f9-b18887b2648c
## INFO [05:21:48.627] [bbotk] Evaluating 1 configuration(s)
## INFO [05:21:49.543] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:21:49.672] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:22:10.626] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:22:30.834] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:22:49.806] [mlr3] Finished benchmark
## INFO [05:22:50.037] [bbotk] Result of batch 3:
## INFO [05:22:50.048] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:22:50.048] [bbotk] 1.343333 95.77653 0 0 59.589
## INFO [05:22:50.048] [bbotk] uhash
## INFO [05:22:50.048] [bbotk] 558cb2f2-b7bb-422e-8ac1-427e3e49bb18
## INFO [05:22:50.054] [bbotk] Evaluating 1 configuration(s)
## INFO [05:22:50.407] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:22:50.452] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:23:05.587] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:23:18.690] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:23:34.830] [mlr3] Finished benchmark
## INFO [05:23:35.250] [bbotk] Result of batch 4:
## INFO [05:23:35.267] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:23:35.267] [bbotk] 0.6866667 95.76875 0 0 44.195
## INFO [05:23:35.267] [bbotk] uhash
## INFO [05:23:35.267] [bbotk] 44c44149-cddc-4825-b7d7-9d4527fc233a
## INFO [05:23:35.280] [bbotk] Evaluating 1 configuration(s)
## INFO [05:23:35.906] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:23:36.145] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:23:54.899] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:24:10.779] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:24:29.823] [mlr3] Finished benchmark
## INFO [05:24:30.204] [bbotk] Result of batch 5:
## INFO [05:24:30.223] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:24:30.223] [bbotk] 0.4677778 95.7672 0 0 53.317
## INFO [05:24:30.223] [bbotk] uhash
## INFO [05:24:30.223] [bbotk] 2e53ea80-2c3d-4bc9-a847-3273f0a8d963
## INFO [05:24:30.241] [bbotk] Evaluating 1 configuration(s)
## INFO [05:24:30.682] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:24:30.773] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:24:51.202] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:25:07.990] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:25:23.056] [mlr3] Finished benchmark
## INFO [05:25:23.192] [bbotk] Result of batch 6:
## INFO [05:25:23.196] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:25:23.196] [bbotk] 1.124444 95.77348 0 0 51.945
## INFO [05:25:23.196] [bbotk] uhash
## INFO [05:25:23.196] [bbotk] 96d932a6-0d7d-42bb-9d4f-2a5442f4952c
## INFO [05:25:23.198] [bbotk] Evaluating 1 configuration(s)
## INFO [05:25:23.302] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:25:23.321] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:25:38.360] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:25:52.365] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:26:07.488] [mlr3] Finished benchmark
## INFO [05:26:07.625] [bbotk] Result of batch 7:
## INFO [05:26:07.630] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:26:07.630] [bbotk] 1.781111 95.78376 0 0 43.994
## INFO [05:26:07.630] [bbotk] uhash
## INFO [05:26:07.630] [bbotk] 45f0510d-f64c-497e-8239-09f680c09dd5
## INFO [05:26:07.633] [bbotk] Evaluating 1 configuration(s)
## INFO [05:26:07.719] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:26:07.740] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:26:44.946] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:26:59.837] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:27:17.904] [mlr3] Finished benchmark
## INFO [05:27:18.269] [bbotk] Result of batch 8:
## INFO [05:27:18.284] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:27:18.284] [bbotk] 0.9055556 95.77086 0 0 69.979
## INFO [05:27:18.284] [bbotk] uhash
## INFO [05:27:18.284] [bbotk] b3f0d7fd-a55b-4f2d-9cfb-01e3443067ef
## INFO [05:27:18.291] [bbotk] Evaluating 1 configuration(s)
## INFO [05:27:18.573] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:27:18.611] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:27:34.863] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:27:50.554] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:28:10.164] [mlr3] Finished benchmark
## INFO [05:28:10.575] [bbotk] Result of batch 9:
## INFO [05:28:10.587] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:28:10.587] [bbotk] 0.2488889 95.76631 0 0 51.38
## INFO [05:28:10.587] [bbotk] uhash
## INFO [05:28:10.587] [bbotk] eb30c4ca-6bee-4978-b5d1-cc32112d496d
## INFO [05:28:10.593] [bbotk] Evaluating 1 configuration(s)
## INFO [05:28:10.910] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:28:10.950] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 2/3)
## INFO [05:28:26.604] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 3/3)
## INFO [05:28:42.884] [mlr3] Applying learner 'scale.imputemean.regr.glmnet' on task 'Delay' (iter 1/3)
## INFO [05:28:57.596] [mlr3] Finished benchmark
## INFO [05:28:57.810] [bbotk] Result of batch 10:
## INFO [05:28:57.817] [bbotk] regr.glmnet.lambda regr.mse warnings errors runtime_learners
## INFO [05:28:57.817] [bbotk] 1.562222 95.77998 0 0 46.518
## INFO [05:28:57.817] [bbotk] uhash
## INFO [05:28:57.817] [bbotk] a5491f00-f8a4-484e-8967-e9d3186d5d8e
## INFO [05:28:57.886] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [05:28:57.888] [bbotk] Result:
## INFO [05:28:57.895] [bbotk] regr.glmnet.lambda learner_param_vals x_domain regr.mse
## INFO [05:28:57.895] [bbotk] 0.03 <list[3]> <list[1]> 95.76616
## INFO [05:29:14.478] [mlr3] Applying learner 'scale.imputemean.regr.ranger.tuned' on task 'Delay' (iter 1/3)
## INFO [05:29:15.310] [bbotk] Starting to optimize 1 parameter(s) with '<TunerGridSearch>' and '<TerminatorEvals> [n_evals=20, k=0]'
## INFO [05:29:15.332] [bbotk] Evaluating 1 configuration(s)
## INFO [05:29:15.492] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:29:15.537] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:29:32.794] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:29:46.953] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:30:03.771] [mlr3] Finished benchmark
## INFO [05:30:04.023] [bbotk] Result of batch 1:
## INFO [05:30:04.037] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:30:04.037] [bbotk] 50 101.6354 0 0 48.047
## INFO [05:30:04.037] [bbotk] uhash
## INFO [05:30:04.037] [bbotk] 2e7f84af-5545-4841-b21e-f1eba4afd333
## INFO [05:30:04.048] [bbotk] Evaluating 1 configuration(s)
## INFO [05:30:04.259] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:30:04.289] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:30:23.210] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:30:38.516] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:30:55.942] [mlr3] Finished benchmark
## INFO [05:30:56.208] [bbotk] Result of batch 2:
## INFO [05:30:56.216] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:30:56.216] [bbotk] 111 100.8052 0 0 51.547
## INFO [05:30:56.216] [bbotk] uhash
## INFO [05:30:56.216] [bbotk] 72c26eaa-3fc7-4953-b6d0-68b36c71c4b0
## INFO [05:30:56.223] [bbotk] Evaluating 1 configuration(s)
## INFO [05:30:56.388] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:30:56.411] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:31:17.292] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:31:37.714] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:31:59.639] [mlr3] Finished benchmark
## INFO [05:31:59.879] [bbotk] Result of batch 3:
## INFO [05:31:59.890] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:31:59.890] [bbotk] 294 100.784 0 0 63.093
## INFO [05:31:59.890] [bbotk] uhash
## INFO [05:31:59.890] [bbotk] 69f34ed5-f41f-448c-b50e-3bc0c5fefb81
## INFO [05:31:59.901] [bbotk] Evaluating 1 configuration(s)
## INFO [05:32:00.103] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:32:00.137] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:32:25.782] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:32:50.133] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:33:15.679] [mlr3] Finished benchmark
## INFO [05:33:15.781] [bbotk] Result of batch 4:
## INFO [05:33:15.785] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:33:15.785] [bbotk] 539 100.8647 0 0 75.435
## INFO [05:33:15.785] [bbotk] uhash
## INFO [05:33:15.785] [bbotk] d9ee3dca-ea30-486a-ba47-862ad1ed8738
## INFO [05:33:15.788] [bbotk] Evaluating 1 configuration(s)
## INFO [05:33:15.910] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:33:15.927] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:33:37.756] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:34:00.712] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:34:22.605] [mlr3] Finished benchmark
## INFO [05:34:22.694] [bbotk] Result of batch 5:
## INFO [05:34:22.699] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:34:22.699] [bbotk] 417 100.505 0 0 66.551
## INFO [05:34:22.699] [bbotk] uhash
## INFO [05:34:22.699] [bbotk] f8d08cb0-8bd3-4442-844e-9826835e1433
## INFO [05:34:22.702] [bbotk] Evaluating 1 configuration(s)
## INFO [05:34:22.821] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:34:22.838] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:34:44.174] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:35:04.689] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:35:26.417] [mlr3] Finished benchmark
## INFO [05:35:26.636] [bbotk] Result of batch 6:
## INFO [05:35:26.643] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:35:26.643] [bbotk] 356 100.5993 0 0 63.5
## INFO [05:35:26.643] [bbotk] uhash
## INFO [05:35:26.643] [bbotk] 2c5ff1a9-a22b-430c-9b16-bceed6c0cffc
## INFO [05:35:26.648] [bbotk] Evaluating 1 configuration(s)
## INFO [05:35:26.817] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:35:26.846] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:35:50.178] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:36:13.301] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:36:37.225] [mlr3] Finished benchmark
## INFO [05:36:37.341] [bbotk] Result of batch 7:
## INFO [05:36:37.348] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:36:37.348] [bbotk] 478 100.8171 0 0 70.293
## INFO [05:36:37.348] [bbotk] uhash
## INFO [05:36:37.348] [bbotk] 31b8b2e3-9fa5-4173-b44d-5b7c41ef46ea
## INFO [05:36:37.353] [bbotk] Evaluating 1 configuration(s)
## INFO [05:36:37.481] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:36:37.503] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:36:54.885] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:37:10.624] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:37:29.015] [mlr3] Finished benchmark
## INFO [05:37:29.190] [bbotk] Result of batch 8:
## INFO [05:37:29.197] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:37:29.197] [bbotk] 172 100.5946 0 0 51.418
## INFO [05:37:29.197] [bbotk] uhash
## INFO [05:37:29.197] [bbotk] 103cdefa-9d5c-41bf-8f15-7118b32db48c
## INFO [05:37:29.202] [bbotk] Evaluating 1 configuration(s)
## INFO [05:37:29.364] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:37:29.385] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:37:46.916] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:38:05.774] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:38:24.240] [mlr3] Finished benchmark
## INFO [05:38:24.305] [bbotk] Result of batch 9:
## INFO [05:38:24.307] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:38:24.307] [bbotk] 233 100.9425 0 0 54.744
## INFO [05:38:24.307] [bbotk] uhash
## INFO [05:38:24.307] [bbotk] 3eb3b18c-5c06-4d9c-8f1b-3ea2555324f6
## INFO [05:38:24.309] [bbotk] Evaluating 1 configuration(s)
## INFO [05:38:24.354] [mlr3] Running benchmark with 3 resampling iterations
## INFO [05:38:24.366] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 2/3)
## INFO [05:38:50.036] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 1/3)
## INFO [05:39:16.375] [mlr3] Applying learner 'scale.imputemean.regr.ranger' on task 'Delay' (iter 3/3)
## INFO [05:39:44.380] [mlr3] Finished benchmark
## INFO [05:39:44.458] [bbotk] Result of batch 10:
## INFO [05:39:44.462] [bbotk] regr.ranger.num.trees regr.mse warnings errors runtime_learners
## INFO [05:39:44.462] [bbotk] 600 100.9782 0 0 79.936
## INFO [05:39:44.462] [bbotk] uhash
## INFO [05:39:44.462] [bbotk] 7129e28c-6760-4d24-80a2-11c73f84b2fb
## INFO [05:39:44.476] [bbotk] Finished optimizing after 10 evaluation(s)
## INFO [05:39:44.477] [bbotk] Result:
## INFO [05:39:44.480] [bbotk] regr.ranger.num.trees learner_param_vals x_domain regr.mse
## INFO [05:39:44.480] [bbotk] 417 <list[3]> <list[1]> 100.505
## INFO [05:40:14.507] [mlr3] Finished benchmarkautoplot(bmr) +
theme_bw() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) 
bmr$aggregate(measure)## nr resample_result task_id learner_id
## 1: 1 <ResampleResult[22]> Delay imputemean.regr.lm
## 2: 2 <ResampleResult[22]> Delay scale.imputemean.regr.glmnet
## 3: 3 <ResampleResult[22]> Delay scale.imputemean.regr.glmnet.tuned
## 4: 4 <ResampleResult[22]> Delay scale.imputemean.regr.ranger.tuned
## resampling_id iters regr.mse
## 1: cv 3 94.54240
## 2: cv 3 94.54240
## 3: cv 3 94.54238
## 4: cv 3 99.45913#nr resample_result task_id
#1: 1 <ResampleResult[22]> Delay
#2: 2 <ResampleResult[22]> Delay
#3: 3 <ResampleResult[22]> Delay
#4: 4 <ResampleResult[22]> Delay
#learner_id resampling_id iters
#1: imputemean.regr.lm cv 3
#2: scale.imputemean.regr.glmnet cv 3
#3: scale.imputemean.regr.glmnet.tuned cv 3
#4: scale.imputemean.regr.ranger.tuned cv 3
#regr.mse
#1: 68.32703
#2: 68.32712
#3: 68.32669
#4: 31.95522Model conclusion:
Random Forests performed the best over the other regression models with an MSE of 31.955.